{
  "version": 3,
  "sources": ["ssg:https://framerusercontent.com/modules/YVgynkVt96a0ES9Gk8yz/nydPZLjpMINuRfzdnMpV/eo4RAmtig-32.js"],
  "sourcesContent": ["import{jsx as e,jsxs as t}from\"react/jsx-runtime\";import{Link as a}from\"framer\";import{motion as o}from\"framer-motion\";import*as n from\"react\";export const richText=/*#__PURE__*/t(n.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"What happens when synonyms are synonyms...until they aren't? One of our clients is a well-known voice assistant, and they came to us with a synonym project. As we look back, we'd like to dive into why the voice assistant needed to tell similar words apart and how we structured the task for Tolokers. \"}),/*#__PURE__*/e(\"h2\",{children:\"How it went down\"}),/*#__PURE__*/e(\"p\",{children:\"The client asked our team of Crowd Solution Architects to set up and manage the entire project from start to finish. Their analysts handed us a dataset with 372,000 word pairs we needed to mark as synonymous or non-synonymous. For Tolokers, that's a piece of cake. Here's what the task looked like for Tolokers:\"}),/*#__PURE__*/e(\"img\",{alt:\"How it went down\",className:\"framer-image\",height:\"694\",src:\"https://framerusercontent.com/images/mHXEiFUJ1ZVSLMpL6oF7FO4KGw.jpeg\",srcSet:\"https://framerusercontent.com/images/mHXEiFUJ1ZVSLMpL6oF7FO4KGw.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/mHXEiFUJ1ZVSLMpL6oF7FO4KGw.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/mHXEiFUJ1ZVSLMpL6oF7FO4KGw.jpeg?scale-down-to=2048 2048w,https://framerusercontent.com/images/mHXEiFUJ1ZVSLMpL6oF7FO4KGw.jpeg 2144w\",style:{aspectRatio:\"2144 / 1388\"},width:\"1072\"}),/*#__PURE__*/e(\"p\",{children:\"For each pair of words, Tolokers needed to mark Yes if they thought the pair was synonymous or No if they didn't. We used an overlap of 5 to optimize quality on this task (each pair of words was shown to five different people and the results were aggregated).\"}),/*#__PURE__*/e(\"p\",{children:\"A quick dig through our textbooks from school told us that two words are synonyms if they are:\"}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Close in meaning.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Different in their spelling and pronunciation.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"The same part of speech.\"})})]}),/*#__PURE__*/e(\"p\",{children:\"The instructions were ready, we had examples for the tutorial and a strict test, and a set of 1000 pairs was primed for sample markup to make sure the client was happy with our quality level. Back then, we naively assumed this was simple and straightforward.\"}),/*#__PURE__*/e(\"h2\",{children:\"Looks great, just do it all over again\"}),/*#__PURE__*/t(\"p\",{children:[\"When we shipped the test results off to the client, it turned out that we had our work cut out for us. For instance, we'd marked \",/*#__PURE__*/e(\"em\",{children:\"red\"}),\" and \",/*#__PURE__*/e(\"em\",{children:\"crimson\"}),\" as synonymous. The client disagreed. From a purely linguistic standpoint (and according to a quick internet search), we were right. The two words refer to very similar shades, which means they're close in meaning. But the client then filled us in on a little detail: while we were taking a big-picture view and looking for similarities, we were supposed to be finding exact synonyms. Not just in the same ballpark, pairs were synonymous if they could be used interchangeably in a variety of contexts. Think: pail and bucket or sick and ill.\"]}),/*#__PURE__*/t(\"p\",{children:[\"That narrowed down the task: find the pairs that are exact synonyms. \\\"Okay, now it's going to be smooth sailing,\\\" we thought once more. But it turned out that even complete similarity isn't enough to guarantee interchangeability. To take one example, words could be from different periods in the language's historical development - \",/*#__PURE__*/e(\"em\",{children:\"twain\"}),\" and \",/*#__PURE__*/e(\"em\",{children:\"two\"}),\", say. Languages are constantly evolving, new contexts for words appear, some words fade into the past, and others are introduced. While \",/*#__PURE__*/e(\"em\",{children:\"twain\"}),\" and \",/*#__PURE__*/e(\"em\",{children:\"two\"}),\" are synonymous, they can only rarely be used interchangeably.\"]}),/*#__PURE__*/e(\"p\",{children:\"The problem is that languages are living, breathing organisms. Finding exact synonyms is a challenge. Even when words appear to have identical meanings, they might drift away from each other over time as stylistic differences beget definitional differences.\"}),/*#__PURE__*/e(\"h2\",{children:\"Houston, what do we do now?\"}),/*#__PURE__*/e(\"p\",{children:\"The problem couldn't have been more foundational. We went back to the client and asked them what business need we were meeting. As it turned out, they were building a tool that writes rules for parsing intents , which are user queries that might have synonyms in them. When a user says, \\\"Call mommy,\\\" the voice assistant is supposed to know that they're talking about their mother. If they're talking about what color their new \\\"ride\\\" is, it's their car.\"}),/*#__PURE__*/e(\"p\",{children:\"Filling us in on the business logic told us what the markup needed to look like. It wasn't exact linguistic synonyms we were looking for. We needed to pick out the synonyms users see as exact matches. In other words, word pairs that are interchangeable in most contexts. Tolokers had new definitions to work with. For our project, two words were synonymous if they were:\"}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"The same part of speech.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Different in their spelling and pronunciation.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Not just similar in meaning, but interchangeable in most contexts.\"})})]}),/*#__PURE__*/t(\"p\",{children:[\"To make things simpler for Tolokers, we came up with a rule: \",/*#__PURE__*/e(\"em\",{children:\"If you can think of multiple contexts where these words can stand in for each other without changing the meaning, they're synonymous\"}),\".\"]}),/*#__PURE__*/e(\"p\",{children:\"The test for Tolokers featured clear-cuts examples without any ambiguity, while the control tasks used to filter out cheaters had a few word pairs that were written and pronounced identically as well as some that were different parts of speech. The threshold for entry was set very high. Given that we were looking to develop a strong linguistic intuition, we needed the same from our Tolokers.\"}),/*#__PURE__*/e(\"h2\",{children:\"Bringing it home\"}),/*#__PURE__*/e(\"p\",{children:\"With the markup complete, we submitted it to the client. And we can draw some conclusions even though the data hasn't hit the production phase yet.\"}),/*#__PURE__*/e(\"p\",{children:\"This is how we rated synonyms based on the results from Tolokers:\"}),/*#__PURE__*/e(\"img\",{alt:\"Bringing it home\",className:\"framer-image\",height:\"761\",src:\"https://framerusercontent.com/images/rhtn3aTGwauxYFTUp79I0iNI.webp\",srcSet:\"https://framerusercontent.com/images/rhtn3aTGwauxYFTUp79I0iNI.webp?scale-down-to=512 512w,https://framerusercontent.com/images/rhtn3aTGwauxYFTUp79I0iNI.webp?scale-down-to=1024 1024w,https://framerusercontent.com/images/rhtn3aTGwauxYFTUp79I0iNI.webp?scale-down-to=2048 2048w,https://framerusercontent.com/images/rhtn3aTGwauxYFTUp79I0iNI.webp 2560w\",style:{aspectRatio:\"2560 / 1522\"},width:\"1280\"}),/*#__PURE__*/e(\"p\",{children:\"The numbers in the image show how consistent Tolokers were in their assessments for each word pair. Each pair was shown to five Tolokers. \"}),/*#__PURE__*/e(\"p\",{children:\"1.0 - All five people thought the two words were synonymous.\"}),/*#__PURE__*/e(\"p\",{children:\"0.8 - Four of the five thought the two words were synonymous.\"}),/*#__PURE__*/e(\"p\",{children:\"0.6 - Here, we have to suspect the words are not synonymous.\"}),/*#__PURE__*/e(\"p\",{children:\"The client walked away happy with our quality level. For our part, we confirmed the importance of starting the project with clear instructions that explain how the results will be used. To achieve high-quality results in fewer iterations, this step is just as essential as adding control tasks and overlap.\"})]});export const richText1=/*#__PURE__*/t(n.Fragment,{children:[/*#__PURE__*/e(\"h2\",{children:\"Why diverse and global crowd makes for better AI\"}),/*#__PURE__*/t(\"p\",{children:[\"The world of machine learning is realizing the need for data collected from the widest possible demographic. Otherwise, the algorithms learn to mimic the biases, prejudices and culturally-specific understandings of a narrow group of people. Some businesses have even built products that \",/*#__PURE__*/e(a,{href:\"https://www.theverge.com/2019/1/25/18197137/amazon-rekognition-facial-recognition-bias-race-gender\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"work differently based on gender and skin color\"})}),\", compromising both the ethics and the effectiveness of AI. \"]}),/*#__PURE__*/t(\"p\",{children:[\"Alternatively, the algorithms that get this right work in more places, deliver more value and serve a global market. Whether it's for data collection, labeling, or building \",/*#__PURE__*/e(a,{href:\"https://medium.com/toloka/automate-that-how-handl-is-saving-companies-money-and-time-on-digitizing-paperwork-213218d18d90\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"human-in-the-loop pipelines\"})}),\", more and more engineers are coming to understand the need for a truly universal human perspective. That's why they want training data labeled by the widest possible mix of genders, nationalities, religions, social and working backgrounds.\"]}),/*#__PURE__*/e(\"p\",{children:\"That's why they want training data labeled by the most diverse crowd available.\"}),/*#__PURE__*/e(\"p\",{children:\"That's why they want training data labeled by Toloka.\"}),/*#__PURE__*/e(\"p\",{children:\"Let's take a closer look at the crowd behind it.\"}),/*#__PURE__*/e(\"h2\",{children:\"Tolokers in 2022\"}),/*#__PURE__*/e(\"p\",{children:\"As of March 2022, Toloka has 245,000 monthly active performers worldwide, which is 20% more than last year. Of the world's 217 countries, Toloka has active performers in 123. Tolokers can access the platform interface in the following languages: English, Spanish, French, Vietnamese, Russian, Uzbek, Turkish, Indonesian, German, Italian, Portuguese-Brazilian, Polish, Chinese, Korean, Japanese, Hindi.\"}),/*#__PURE__*/e(\"p\",{children:\"Out of the 10,000 Tolokers surveyed* worldwide, only 34.3% identify as White / Caucasian, whereas 23.7% are Black, 29.0% Asian, and 2.2% Indigenous. Global Tolokers belong to over 600 different ethnicities, from Filipino to Kinh, and from Venezolana to Kikuyu.\"}),/*#__PURE__*/e(\"p\",{children:\"Tolokers are 29.8 years old on average, and their gender distribution is 39.7% female, 58.8% male, 0.8% non-binary. The majority of them are single (63.7%) and have no children (70.7%), presumably due to their young age.\"}),/*#__PURE__*/e(\"h2\",{children:\"Socioeconomic status\"}),/*#__PURE__*/e(\"p\",{children:\"Considerable evidence suggests that socioeconomic status, broadly defined, explains a range of important features, from judgments to life outcomes.\"}),/*#__PURE__*/e(\"p\",{children:\"Tolokers come from countries with wildly differing economies, from Burundi, where the GDP per capita is $274, to Norway, where it's $67,390. Matching this global range, Tolokers' individual income also varies from under $1000 to $100,000 yearly. The global median annual income of Tolokers is about $4000. Most self-identify as working or lower-middle class.\"}),/*#__PURE__*/e(\"h2\",{children:\"Education and occupation\"}),/*#__PURE__*/e(\"p\",{children:\"Global Tolokers appear well-educated: 82.9% report finishing at least some college, and 4.7% have advanced degrees. 72% report intermediate to advanced English skills (Toloka verifies performers' language skills by the platform quality control tools). Tolokers work in many industries, from service to manufacturing, and from healthcare to finance.\"}),/*#__PURE__*/e(\"h2\",{children:\"Religion and culture\"}),/*#__PURE__*/e(\"p\",{children:\"66.9% of Tolokers identify themselves as religious or spiritual. Tolokers represent a diverse spectrum of world religions, from Judaism to Shint\u014D, and from Muism to Cao Dai, most of them observing Islam (46.6%), Christianity (40.4%), Hinduism (7.3%), and Buddhism (2.4%).\"}),/*#__PURE__*/e(\"p\",{children:\"Tolokers speak a variety of languages including exotic ones such as Namwanga, Ikwerre, Hiligaynon, or Ewe; in total, they report speaking over 800 different languages and dialects. Most Tolokers (98%) speak at least some English, and many of them also know Spanish (39%), French (36%), Arabic (31%), Portuguese (22%), German (21%), Turkish (21%), Italian (18%), and Russian (11%).\"}),/*#__PURE__*/e(\"p\",{children:\"Among their hobbies, Tolokers mention sports, reading, computer games, music, cinema, and... Toloka!\"}),/*#__PURE__*/e(\"h2\",{children:\"COVID-19\"}),/*#__PURE__*/e(\"p\",{children:\"When the global pandemic struck, we saw an influx of new users to Toloka. Since then, many have reported that Toloka has become a vital source of additional income during the pandemic. In this year's survey, 75% of Tolokers reported they'd lost at least some income during the pandemic years, so Toloka keeps providing important additional income to disadvantaged populations around the globe.\"}),/*#__PURE__*/e(\"h2\",{children:\"Why they do labeling\"}),/*#__PURE__*/t(\"p\",{children:[\"Is Toloka a main source of income for its users? Mostly not. 65.8% of Tolokers are permanently employed and 60.5% report having stable income. They use \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/knowledgebase/crowdsourcing-concepts/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"crowdsourcing\"})}),\" for reasons other than breadwinning.\"]}),/*#__PURE__*/e(\"p\",{children:\"Actually, that's exactly what they say: many Tolokers report doing microwork for reasons other than earning money, such as: for fun (17%), to make the world a better place (17%), to gain work experience and advance career (13%), or to be part of modern technology (11%). However, earning extra cash is on the table, too: 62% Tolokers mainly use Toloka as an additional source of income, while for 18% of performers Toloka is the main source of one.\"}),/*#__PURE__*/t(\"p\",{children:[\"Tolokers also just enjoy \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/data-labeling-platform/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"data labeling\"})}),'. 81% anonymously report they love Toloka. Among reasons, they mention Toloka\\'s flexibility (\"you can pick what you like\"; \"even when I take care of my child at home I can earn to help my spouse with the everyday expenses\") and learning opportunities (\"it stimulates my visual, logical, and intellectual capacity\", \"helps me practice foreign languages\", \"with Toloka, I can learn and earn\"). Some Tolokers confess that Toloka refreshes them on school lectures and helps them kill time. Generally, Tolokers seem to share our own belief that the platform is \"good for both people and the improving technologies.\"']}),/*#__PURE__*/e(\"p\",{children:\"As algorithms take over more and more meaningful decisions, it matters more and more that these algorithms learn from high-quality, diverse, and representative data. With Toloka you can harvest quality training data with a truly global perspective, and do so responsibly. \"}),/*#__PURE__*/e(\"img\",{alt:\"This will only become more important\",className:\"framer-image\",height:\"233\",src:\"https://framerusercontent.com/images/aVvzXRyYfImnatbYg7adLeeEeo.png\",srcSet:\"https://framerusercontent.com/images/aVvzXRyYfImnatbYg7adLeeEeo.png?scale-down-to=512 512w,https://framerusercontent.com/images/aVvzXRyYfImnatbYg7adLeeEeo.png?scale-down-to=1024 1024w,https://framerusercontent.com/images/aVvzXRyYfImnatbYg7adLeeEeo.png 1888w\",style:{aspectRatio:\"1888 / 466\"},width:\"944\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"br\",{className:\"trailing-break\"})})]});export const richText2=/*#__PURE__*/t(n.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"Competitor pricing information is a strategic weapon in e-commerce \u2014 and inaccurate information may ultimately lead to revenue loss. With the goal to sharpen their pricing strategy, Yandex.Market turned to Toloka for help with their large-scale price matching. \"}),/*#__PURE__*/e(\"h2\",{children:\"Challenge\"}),/*#__PURE__*/e(\"p\",{children:\"Yandex.Market is a large online marketplace with a huge assortment of products. They need to continually collect information on the catalogs and prices from other major retailers in order to meet the following business goals:\"}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Perform market analysis to learn about competitors\u2019 product assortment and pricing.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Estimate the overlap in product categories.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Analyze competitor prices, find the market equilibrium, and dynamically adjust pricing where necessary.\"})})]}),/*#__PURE__*/t(\"p\",{children:[\"Automated algorithms match items on retailer websites, but they don\u2019t consistently perform well enough to achieve the business outcomes set out by Yandex.Market. Human-\",/*#__PURE__*/e(a,{href:\"https://toloka.ai/data-labeling-platform/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"labeled data\"})}),\" is needed to improve match quality and cover missing matches. For instance, human annotators are better at handling issues like matching identical items that have different names.\"]}),/*#__PURE__*/e(\"p\",{children:\"In pursuit of high-quality data, the company started off with an in-house labeling team, but it proved to be an expensive asset. After considering other available options, they chose Toloka for its quality, speed, and affordability, with an important deciding factor \u2014 the API allowed them to integrate with existing internal pipelines.\"}),/*#__PURE__*/e(\"p\",{children:\"Tasks in Toloka were designed to serve three purposes:\"}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Assess the quality of automatic matching.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Improve the quality of existing matches by removing incorrect matches.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Increase match coverage by finding more URLs of matching items on competitor sites.\"})})]}),/*#__PURE__*/e(\"h2\",{children:\"Solution\"}),/*#__PURE__*/e(\"p\",{children:\"When Toloka stepped in, there were two task components for the crowd performers to tackle:\"}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Find and save a URL link \u2013 Tolokers identify specific products on various e-commerce websites.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Check and compare \u2013 Tolokers decide whether a pair of products with different URLs are the same. \"})})]}),/*#__PURE__*/e(\"img\",{alt:\"solution\",className:\"framer-image\",height:\"706\",src:\"https://framerusercontent.com/images/XNS3WNtUD1iJajkfrU2vxZgJs.webp\",srcSet:\"https://framerusercontent.com/images/XNS3WNtUD1iJajkfrU2vxZgJs.webp?scale-down-to=512 512w,https://framerusercontent.com/images/XNS3WNtUD1iJajkfrU2vxZgJs.webp?scale-down-to=1024 1024w,https://framerusercontent.com/images/XNS3WNtUD1iJajkfrU2vxZgJs.webp?scale-down-to=2048 2048w,https://framerusercontent.com/images/XNS3WNtUD1iJajkfrU2vxZgJs.webp 2560w\",style:{aspectRatio:\"2560 / 1413\"},width:\"1280\"}),/*#__PURE__*/t(\"p\",{children:[\"An efficient pipeline was designed to be compatible with the company\u2019s internal processes. The pipeline includes two directions of interaction, from the dynamic pricing system to Toloka and back, with \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/data-labeling-platform/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"data labeling\"})}),\" in the middle.\"]}),/*#__PURE__*/e(\"img\",{alt:\"dynamic pricing system to Toloka\",className:\"framer-image\",height:\"290\",src:\"https://framerusercontent.com/images/98jK3sV2ny0K2HK4szB965UZjg.jpeg\",srcSet:\"https://framerusercontent.com/images/98jK3sV2ny0K2HK4szB965UZjg.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/98jK3sV2ny0K2HK4szB965UZjg.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/98jK3sV2ny0K2HK4szB965UZjg.jpeg 1840w\",style:{aspectRatio:\"1840 / 580\"},width:\"920\"}),/*#__PURE__*/e(\"p\",{children:\"Four major steps are embedded within the pipeline:\"}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"The preparatory stage\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Data collection\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Quality control\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Labeling and accuracy check\"})})]}),/*#__PURE__*/e(\"p\",{children:\"The company uses automated pre-labeling to prepare and verify each pool of URL links before sending them to Toloka for human labeling. All outdated and visibly erroneous matches are removed. The remaining links with potential matches are left for Tolokers to analyze. \"}),/*#__PURE__*/e(\"h2\",{children:\"Results\"}),/*#__PURE__*/e(\"p\",{children:\"The Toloka pipeline provides 2.5% better coverage of key products compared to using automated solutions and in-house annotators. Tolokers are used on demand to find competitor products in priority categories \u2014 groups of products that contribute the most to GMV and have the greatest business value for Yandex.Market. As the project is ongoing, labeling accuracy and speed continue to improve as Tolokers build their skills. To maintain quality, the project\u2019s honeypots (control tasks) are updated regularly to make sure that the URL links are active and all of the listed products remain in stock.\"})]});export const richText3=/*#__PURE__*/t(n.Fragment,{children:[/*#__PURE__*/t(\"p\",{children:[\"Toloka processes huge amounts of data every day, and protecting this data lies at the core of every update we put out. We\u2019re proud to announce that after a series of in-depth audits, the Toloka platform has obtained \",/*#__PURE__*/e(a,{href:\"https://toloka-cdn.azureedge.net/security-docs/IS748118.pdf\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"certification\"})}),\" for compliance with ISO 27001:2013, the highest industry standard in information security management. \"]}),/*#__PURE__*/e(\"h2\",{children:\"What is ISO 27001:2013?\"}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(a,{href:\"https://www.iso.org/isoiec-27001-information-security.html\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"ISO/IEC 27001\"})}),\" is an international information security certificate originally introduced in 2005 and then revised in 2013. It\u2019s jointly awarded by the International Organization for Standardization \",/*#__PURE__*/e(a,{href:\"https://www.iso.org/about-us.html\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"(ISO)\"})}),\" and the International Electrotechnical Commission \",/*#__PURE__*/e(a,{href:\"https://www.iec.ch/about-us\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"(IEC)\"})}),\". Both are independent, non-governmental organizations based in Switzerland.\"]}),/*#__PURE__*/e(\"p\",{children:\"Companies that obtain ISO 27001:2013 are compliant with the highest information security management system (ISMS) standards published by ISO and IEC. This confirms that an ISO 27001-certified business (along with its products) follows established best practices to protect client data, manage the security of all assets, and reduce any potential security breaches. This covers several key aspects of data storage and maintenance, both within data centers and in the cloud, including but not limited to:\"}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"legal documents\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"financial information\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"intellectual property\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"personal data\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"confidential information entrusted by third parties\"})})]}),/*#__PURE__*/e(\"h2\",{children:\"Getting the certificate\"}),/*#__PURE__*/t(\"p\",{children:[\"While the ISMS protocol lists the standards put in place by ISO and IEC, any company wishing to obtain ISO 27001 needs to be audited at length. In the case of Toloka, the auditor was BSI \u2013 an internationally recognized London-based standards body and certification company. BSI is accredited by \",/*#__PURE__*/e(a,{href:\"https://anab.ansi.org/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"ANAB\"})}),\", a board of the American National Standards Institute \",/*#__PURE__*/e(a,{href:\"https://www.ansi.org/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"(ANSI)\"})}),\", which is a member of the International Accreditation Forum \",/*#__PURE__*/e(a,{href:\"https://iaf.nu/en/home/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"(IAF)\"})}),\".\"]}),/*#__PURE__*/e(\"p\",{children:\"To get the green light from BSI, Toloka went through a series of extensive checks that examined all of the company\u2019s internal processes related to data management and information security. These audits will continue on a regular basis in order to renew the certificate periodically \u2014 and to ensure compliance with the highest industry standards in the future.\"}),/*#__PURE__*/e(\"h2\",{children:\"Additional information\"}),/*#__PURE__*/t(\"p\",{children:[\"To learn more about privacy and information security at Toloka, please \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/security\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"visit this page\"})}),\" \u2013 it outlines Toloka\u2019s commitment to security, explains our legal policies, and provides additional information about our compliances and certificates.\"]})]});export const richText4=/*#__PURE__*/t(n.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"Content moderation is the job of monitoring user-generated web content to make sure that nothing inappropriate, offensive, or disturbing gets published online. Things like false advertising, fraud, or sexually explicit images can seriously harm the reputation of an online business, whether it\u2019s an e-commerce marketplace, an AdTech platform, or a social media site. \"}),/*#__PURE__*/t(\"p\",{children:[\"As online content grows exponentially, keeping the web safe is no easy task. Companies today use Machine Learning models to build their automated moderation systems. These systems rely heavily on high-quality, accurately \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/data-labeling-platform/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"labeled data\"})}),\" for training and validating their models \u2014 and this is where \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/knowledgebase/crowdsourcing-concepts/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"crowdsourcing\"})}),\" can play an invaluable role.\"]}),/*#__PURE__*/e(\"h2\",{children:\"Use case: Yandex Zen\"}),/*#__PURE__*/e(\"p\",{children:\"Yandex Zen is a popular recommendation service that provides user-generated content and uses ML algorithms to support automated moderation. The service takes the form of a personalized news feed that shows articles and videos from external digital media sources, as well as original pieces from its own blog. The website has 20 million daily active users, including 5000 daily new authors, the most popular of whom are able to monetize their contributions.\"}),/*#__PURE__*/e(\"img\",{alt:\"Examples of content on Yandex Zen\",className:\"framer-image\",height:\"1285\",src:\"https://framerusercontent.com/images/RD4YEkwfxjoNVRcLKNXIKZbsfks.webp\",srcSet:\"https://framerusercontent.com/images/RD4YEkwfxjoNVRcLKNXIKZbsfks.webp?scale-down-to=1024 1019w,https://framerusercontent.com/images/RD4YEkwfxjoNVRcLKNXIKZbsfks.webp?scale-down-to=2048 2039w,https://framerusercontent.com/images/RD4YEkwfxjoNVRcLKNXIKZbsfks.webp 2560w\",style:{aspectRatio:\"2560 / 2571\"},width:\"1280\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Examples of content on Yandex Zen \"})}),/*#__PURE__*/e(\"p\",{children:\"Since its launch in 2015, Yandex Zen has encountered 2 major issues:\"}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Even relevant, personalized content on the platform sometimes crosses the line between the acceptable and the inappropriate, going from informative to provocative, sensational, and sometimes even potentially dangerous advice.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Profit-seeking authors try to cheat the website\u2019s algorithm, creating questionable or substandard content in an attempt to increase engagement.\"})})]}),/*#__PURE__*/t(\"p\",{children:[\"The platform relies on human moderators to cover the grey areas that their ML algorithms can\u2019t handle reliably, as well as to monitor trending content in real time. The team found it was a challenge to keep enough moderators online 24/7, label enough content to train and validate their \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/adaptive-ml-models/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"ML models\"})}),\", and maintain quality of moderation.\"]}),/*#__PURE__*/e(\"h2\",{children:\"Challenge\"}),/*#__PURE__*/e(\"p\",{children:\"Toloka was asked to utilize the power of the crowd and help improve moderation at Yandex Zen. Moderators were expected to identify the following types of violations:\"}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Clickbaiting\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Hate speech\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Poor (and potentially hazardous) medical advice\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Spam\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Violent content\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"COVID-19 misinformation\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Adult content\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Illegal content and scams (information about illegal products and services + attempts to defraud other users)\"})})]}),/*#__PURE__*/e(\"img\",{alt:\"Example of a content moderation project\",className:\"framer-image\",height:\"413\",src:\"https://framerusercontent.com/images/Z8GwpbHelt6MoJJWh0PBlelBw0o.jpeg\",srcSet:\"https://framerusercontent.com/images/Z8GwpbHelt6MoJJWh0PBlelBw0o.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/Z8GwpbHelt6MoJJWh0PBlelBw0o.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/Z8GwpbHelt6MoJJWh0PBlelBw0o.jpeg 1588w\",style:{aspectRatio:\"1588 / 826\"},width:\"794\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Example of a content moderation project\"})}),/*#__PURE__*/e(\"p\",{children:\"Toloka\u2019s Crowd Solution Architects (CSAs) isolated 3 classification formats for the project \u2013 text, image, and video.\"}),/*#__PURE__*/e(\"p\",{children:\"Normally, majority vote is used to improve quality on classification tasks. This means that each item is classified by multiple Tolokers and results are automatically aggregated to use the most popular verdict. However, this moderation project was a long-running continuous process, and majority vote revealed socio-demographic bias that degraded quality over time.\"}),/*#__PURE__*/e(\"p\",{children:\"In addition, the website\u2019s fast-paced environment with freshly published hourly content meant that moderation and control tasks had to be regular and frequent, while all trending content required real-time moderation.\"}),/*#__PURE__*/e(\"img\",{alt:\"Example of a classification task about traveling to Egypt\",className:\"framer-image\",height:\"490\",src:\"https://framerusercontent.com/images/byikyhtcImjNGFd40eoOMvtxDc.jpeg\",srcSet:\"https://framerusercontent.com/images/byikyhtcImjNGFd40eoOMvtxDc.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/byikyhtcImjNGFd40eoOMvtxDc.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/byikyhtcImjNGFd40eoOMvtxDc.jpeg 1254w\",style:{aspectRatio:\"1254 / 980\"},width:\"627\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Example of a classification task about traveling to Egypt that was classified as \u201CCovid-19 conspiracy\u201D\"})}),/*#__PURE__*/e(\"p\",{children:\"Toloka created 10 projects for articles and videos, as well as 4 projects for the comment section. Each project contained around 10,000 daily crowdsourcing tasks.\"}),/*#__PURE__*/e(\"h2\",{children:\"Solution\"}),/*#__PURE__*/e(\"p\",{children:\"The Yandex Zen team came up with a new infrastructure for moderating content efficiently. Initially, the majority of content is handled by automatic classifiers. Anything that falls into a gray area is passed on to the Toloka crowd for classification. This crowd is supported by in-house moderators, who verify the performers\u2019 work and help control quality. A handful of expert moderators create control tasks with true labels, which are used for daily secret exams to monitor quality and reward those who perform well. In-house moderators can focus on trending content that demands careful consideration, while letting the crowd handle the majority of tasks.\"}),/*#__PURE__*/e(\"img\",{alt:\"The pipeline enabled effective quality control\",className:\"framer-image\",height:\"558\",src:\"https://framerusercontent.com/images/F4UPjBeOwFNaRAXrlwCsm5kZf0.jpeg\",srcSet:\"https://framerusercontent.com/images/F4UPjBeOwFNaRAXrlwCsm5kZf0.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/F4UPjBeOwFNaRAXrlwCsm5kZf0.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/F4UPjBeOwFNaRAXrlwCsm5kZf0.jpeg 2032w\",style:{aspectRatio:\"2032 / 1116\"},width:\"1016\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"The pipeline enabled effective quality control, transparency, and scalability, and also reduced the need for full-time moderators.\"})}),/*#__PURE__*/e(\"p\",{children:\"Here\u2019s what the new collaborative process achieves:\"}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Successful decomposition of a large-scale labeling project into numerous simple tasks, which makes the workload manageable and increases the overall quality of moderation while reducing the burden on in-house moderators.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"A new quality-based payment scheme that awards bonuses for accuracy and motivates both Tolokers and the in-house staff.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"An interactive model that establishes an effective feedback loop among the moderators and the crowd, resulting in higher accuracy and improved task pool management.\"})})]}),/*#__PURE__*/e(\"p\",{children:\"As a result, moderation accuracy has risen significantly, reaching figures up to 98 percent. Content classification quality also improved in the first 2,5 months of working together with Toloka. \"}),/*#__PURE__*/e(\"h2\",{children:\"Results\"}),/*#__PURE__*/e(\"p\",{children:\"As a result, moderation accuracy has risen significantly, reaching figures up to 98 percent. Content classification quality also improved in the first 2,5 months of working together with Toloka. If your business requires an affordable team of labelers for large-scale projects, including but not limited to classification tasks, be sure to get in touch with us \u2013 we'll be glad to help you build a robust data-labeling pipeline.\"})]});export const richText5=/*#__PURE__*/t(n.Fragment,{children:[/*#__PURE__*/t(\"p\",{children:[\"We continue our special series about one of Toloka\u2019s most prized assets \u2013 our invaluable crowd performers, aka Tolokers, whose efforts make \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/data-labeling-platform/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"data labeling\"})}),\" prompt, accurate, and sought-after by our clients. Be sure to check out the first two stories \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/blog/toloker-stories\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"here\"})}),\" and \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/blog/toloker-stories-2\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"here\"})}),\". \"]}),/*#__PURE__*/e(\"p\",{children:\"Today, we\u2019re talking to Andressa from Brazil and Ashish from India who\u2019ve agreed to have a chat with us to offer their perspective on what it\u2019s like to be a Toloker in 2021.\"}),/*#__PURE__*/e(\"h2\",{children:\"Please tell us about yourself\"}),/*#__PURE__*/e(\"p\",{children:\"Andressa: Hi there, I\u2019m Andressa. I\u2019m 32 years old, and I come from Caxias do Sul, Brazil, in the south of the country. I work as an assembler.\"}),/*#__PURE__*/e(\"p\",{children:\"Ashish: Hello, my name\u2019s Ashish. I\u2019m 21, and I come from Gurugram, India, not far away from the Indian capital. I\u2019m a customer service representative.\"}),/*#__PURE__*/e(\"img\",{alt:\"andressa&ashish\",className:\"framer-image\",height:\"908\",src:\"https://framerusercontent.com/images/Ht8hsDnYuI0XMo5oeeSvVVZItbg.webp\",srcSet:\"https://framerusercontent.com/images/Ht8hsDnYuI0XMo5oeeSvVVZItbg.webp?scale-down-to=512 512w,https://framerusercontent.com/images/Ht8hsDnYuI0XMo5oeeSvVVZItbg.webp?scale-down-to=1024 1024w,https://framerusercontent.com/images/Ht8hsDnYuI0XMo5oeeSvVVZItbg.webp?scale-down-to=2048 2048w,https://framerusercontent.com/images/Ht8hsDnYuI0XMo5oeeSvVVZItbg.webp 2560w\",style:{aspectRatio:\"2560 / 1817\"},width:\"1280\"}),/*#__PURE__*/e(\"h2\",{children:\"What are your hobbies and interests? \"}),/*#__PURE__*/e(\"p\",{children:\"Andressa: I love camping, watching films and TV series, playing with my children, going out with friends whenever possible, and shopping, which is something that requires a regular source of income, of course!\"}),/*#__PURE__*/e(\"p\",{children:\"Ashish: I like to code, play computer games, and I also enjoy reading printed books.\"}),/*#__PURE__*/e(\"h2\",{children:\"How did you find out about Toloka and why did you decide to join?\"}),/*#__PURE__*/e(\"p\",{children:\"Andressa: I found out about Toloka through YouTube. The prospect seemed very promising: Toloka looked like a very professional, well-organized platform that offered pretty straightforward tasks along with\u2014what I thought\u2014was very reasonable payment.\"}),/*#__PURE__*/e(\"p\",{children:\"Ashish: I was looking for passive income and stumbled upon a social media channel that talked at length about the advantages of becoming a crowd performer. Since my goal was to start earning extra money by working in my spare time whenever I felt like it, Toloka ticked all of the boxes.\"}),/*#__PURE__*/e(\"h2\",{children:\"When did you start doing your first tasks?\"}),/*#__PURE__*/e(\"p\",{children:\"Andressa: I started doing tasks for Toloka quite recently, at the end of August. I\u2019ve been enjoying it so far!\"}),/*#__PURE__*/e(\"p\",{children:\"Ashish: I started about a year ago. Then I had to take a break from being a crowd performer because I had to finish my studies. And after that, I came back. Things are great at the moment.\"}),/*#__PURE__*/e(\"h2\",{children:\"What types of tasks do you prefer and why?\"}),/*#__PURE__*/e(\"p\",{children:\"Andressa: Thankfully, Toloka\u2019s ecosystem makes it possible to use any device. Since I prefer to use my smartphone as opposed to my laptop, I normally choose assignments that I can do on the go: mostly surveys and field tasks related to my fields of interest.\"}),/*#__PURE__*/e(\"p\",{children:\"Ashish: I pretty much like all of them, or at least most. Generally speaking, the tasks I choose fall into two categories \u2013 those I do online, and those I do outside.\"}),/*#__PURE__*/e(\"p\",{children:\"For the strictly online data-labeling tasks, I use mostly my PC, and sometimes my smartphone. Since I have a good typing speed, I often do online surveys, as well as annotations. And for the outside tasks, it\u2019s just me and my phone \u2013 as easy as that. Because I live in Gurugram, I have access to all sorts of shops, cafes, and other businesses. This makes completing field tasks not only easy, but also enjoyable. I pretty much have a good time being out and about and get paid for it.\"}),/*#__PURE__*/e(\"h2\",{children:\"What unusual tasks, if any, have you completed on Toloka?\"}),/*#__PURE__*/e(\"p\",{children:\"Andressa: Nothing too unusual so far. Everything I\u2019ve been doing makes perfect sense, especially when the instructions are clear.\"}),/*#__PURE__*/e(\"p\",{children:\"Ashish: The more tasks you do, the less surprising they become, and the easier they are to complete. At this point, I can\u2019t really recall any unusual tasks. Perhaps, they were unusual at the time I got them, but they all seem pretty usual to me today.\"}),/*#__PURE__*/e(\"h2\",{children:\"How often do you work and how do you spend the money you earn with Toloka?\"}),/*#__PURE__*/e(\"p\",{children:\"Andressa: It really depends on how busy I get with my main job and also my family. I can do Toloka anywhere from everyday or every other day to as little as once a week. Whatever I make I use towards my household expenses.\"}),/*#__PURE__*/e(\"p\",{children:\"Ashish: 3-4 hours per day, pretty much every day when I have time, or less when I don\u2019t (like when I had to do my college exams). I use the money I make with Toloka to pay for my mobile expenses and/or Internet and also to buy clothes from time to time.\"}),/*#__PURE__*/e(\"h2\",{children:\"What do you like about Toloka (other than making money)?\"}),/*#__PURE__*/e(\"p\",{children:\"Andressa: I like that the whole data-labeling experience is interactive. And I can access Toloka very easily from anywhere in the world. It\u2019s about my freedom of choice and convenience.\"}),/*#__PURE__*/e(\"p\",{children:\"Ashish: Basically, it comes down to the tasks themselves \u2013 I get to choose only what I like, and only when I feel like it. And it doesn\u2019t have to be the same every time. It\u2019s completely up to me, so I feel like I\u2019m in charge of my own schedule and lifestyle.\"}),/*#__PURE__*/e(\"h2\",{children:\"What are the advantages of being a data labeler?\"}),/*#__PURE__*/e(\"p\",{children:\"Andressa: The fact that I learn quite a bit. So, not only is being a labeler financially rewarding, but it\u2019s also very educational. I become more and more knowledgeable, and I feel like I\u2019m improving my expertise. And I also like the fact that the tasks aren\u2019t dull.\"}),/*#__PURE__*/e(\"p\",{children:\"Ashish: Being able to choose tasks that suit my location and mood. I\u2019m not obliged to do anything if I don\u2019t want to do it. I choose only those tasks that I find engaging. And at the same time, I feel like I\u2019m contributing to the world and helping people.\"}),/*#__PURE__*/e(\"h2\",{children:\"What would you tell someone who is thinking about trying Toloka?\"}),/*#__PURE__*/e(\"p\",{children:\"Andressa: I\u2019d say that it\u2019s an exciting decision that you will likely not regret. Unlike many other platforms, everything with Toloka is super simple. Registration is quick, and the platform itself has clearly been built to offer the best user experience, which includes us, Tolokers. I haven\u2019t encountered a single bug so far.\"}),/*#__PURE__*/t(\"p\",{children:[\"Ashish: You should definitely try it \u2013 you\u2019ve got nothing to lose really. It's a legit website that pays fairly for your time and effort, and lots of interesting assignments are available all the time. The only thing I would add is that if you\u2019re totally new to \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/knowledgebase/crowdsourcing-concepts/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"crowdsourcing\"})}),\" and data labeling, you will probably benefit from watching a couple of videos for beginners and doing a little bit of reading beforehand. Toloka has got plenty of these materials, so you won\u2019t need to look far at all.\"]})]});export const richText6=/*#__PURE__*/t(n.Fragment,{children:[/*#__PURE__*/t(\"p\",{children:[\"We're counting down to the \",/*#__PURE__*/e(a,{href:\"https://icml.cc/Conferences/2021\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"International Conference on Machine Learning (ICML)\"})}),' \u2014 the leading global conference in the field of AI and ML. As a Gold Level sponsor of this year\\'s event, Toloka will present a comprehensive 4-hour virtual workshop titled \"High-Quality Data Labeling at Scale with Toloka\" on July 18. ']}),/*#__PURE__*/e(\"h2\",{children:\"About the workshop\"}),/*#__PURE__*/e(\"p\",{children:\"Today, AI development rests on three pillars: algorithms, hardware, and data. For successful advances, the industry needs solutions that can bolster data quality, scalability, and flexibility. Crowdsourcing can feasibly offer these.\"}),/*#__PURE__*/t(\"p\",{children:[\"Our workshop will show you what the Toloka platform has to offer. We'll talk about the infrastructure for data production, practical aspects of designing \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/data-labeling-platform/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"data labeling\"})}),\" tasks, the emerging profession of Crowd Solutions Architect, and the future of work for performers.\"]}),/*#__PURE__*/e(\"h2\",{children:\"Speakers\"}),/*#__PURE__*/e(\"img\",{alt:\"Speakers\",className:\"framer-image\",height:\"981\",src:\"https://framerusercontent.com/images/C5EKLAXkeqW6jGU5bWG6cfNXJU.webp\",srcSet:\"https://framerusercontent.com/images/C5EKLAXkeqW6jGU5bWG6cfNXJU.webp?scale-down-to=512 512w,https://framerusercontent.com/images/C5EKLAXkeqW6jGU5bWG6cfNXJU.webp?scale-down-to=1024 1024w,https://framerusercontent.com/images/C5EKLAXkeqW6jGU5bWG6cfNXJU.webp?scale-down-to=2048 2048w,https://framerusercontent.com/images/C5EKLAXkeqW6jGU5bWG6cfNXJU.webp 2560w\",style:{aspectRatio:\"2560 / 1962\"},width:\"1280\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Speakers\"})}),/*#__PURE__*/t(\"p\",{children:[\"Our lineup of experts will share their practical experience and insights into \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/knowledgebase/crowdsourcing-concepts/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"crowdsourcing\"})}),\":\"]}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(a,{href:\"https://www.linkedin.com/in/omegorskaya/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Olga Megorskaya\"})}),', Toloka\\'s CEO: \"Evolution of Data Production Paradigm in AI. Key Components of Future Success.\" Olga will discuss how AI industry needs can be adequately met during data production and processing. ']})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(a,{href:\"https://www.linkedin.com/in/omar-alonso-8a5235/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Omar Alonso\"})}),', Senior Engineering Manager at Instacart: \"The Practice of Crowdsourcing.\" Omar will discuss practical considerations for designing and completing high-quality tasks that require the combined efforts of both humans and machines.']})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(a,{href:\"https://www.linkedin.com/in/dbaidakova/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Daria Baidakova\"})}),', Director of Educational Programs at Toloka: \"Data Annotation at Scale. A Core Expertise of Modern ML.\" Daria will offer insights into the job of a Crowd Solutions Architect and explain Toloka\\'s research grants program and Crowd Science initiative.']})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(a,{href:\"https://www.linkedin.com/in/saiphcita/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Saiph Savage\"})}),', Assistant Professor at Northeastern University and Co-Director of the Civic Innovation Lab at UNAM: \"The Future of Work for Performers. Empowering the People Behind AI.\" Saiph will propose a framework that facilitates a painless transition to new AI jobs that are unlikely to be automated in the future.']})})]}),/*#__PURE__*/e(\"h2\",{children:\"Demo\"}),/*#__PURE__*/t(\"p\",{children:[\"The keynote part will be followed by a demo that shows how crowdsourcing can tackle an e-commerce item retrieval and ranking task. Software developers, \",/*#__PURE__*/e(a,{href:\"https://www.linkedin.com/in/ustalov/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Dmitry Ustalov\"})}),\" and \",/*#__PURE__*/e(a,{href:\"https://www.linkedin.com/in/vladimir-losev-47318b50/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Vladimir Losev\"})}),\", and Crowd Solutions Architect, \",/*#__PURE__*/e(a,{href:\"https://www.linkedin.com/in/oleg-pavlov-b2766274/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Oleg Pavlov\"})}),\", will join the Toloka team for a hands-on tutorial. The audience will have a chance to learn how to build a human-in-the-loop pipeline from scratch:\"]}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/t(\"p\",{children:[\"Raw crowdsourced data will be integrated into \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/adaptive-ml-models/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"ML models\"})}),\" to obtain a ready-to-use dataset.\"]})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/t(\"p\",{children:[\"The team will show how interdependent data labeling processes can be combined using the \",/*#__PURE__*/e(a,{href:\"https://github.com/Toloka/toloka-kit\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Toloka Kit\"})}),\" (our open-source Python library).\"]})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/t(\"p\",{children:[\"The \",/*#__PURE__*/e(a,{href:\"https://github.com/Toloka/crowd-kit\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Crowd Kit\"})}),\" (quality control library) will be used to complete the task and get clean, AI-ready data.\"]})})]}),/*#__PURE__*/e(\"p\",{children:\"The event will conclude with a Q&A session in real time, during which the team will address all of your questions.\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(a,{href:\"https://toloka.ai/events/workshop-icml2021\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Learn more\"})})})]});export const richText7=/*#__PURE__*/t(n.Fragment,{children:[/*#__PURE__*/t(\"p\",{children:[\"AliExpress has the highest traffic of any e-commerce website in Russia with almost 9 million daily visitors and more than 100 million items for sale. When AliExpress Russia opened its doors in 2019, the company quickly ran into localization problems with many inaccurate translations of catalogued products that required lengthy and ultimately unsuccessful solutions. Looking for new ways to solve this, AliExpress turned to Toloka for \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/knowledgebase/crowdsourcing-concepts/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"crowdsourcing\"})}),\"-based solutions, which proved to be the right decision. \"]}),/*#__PURE__*/e(\"h2\",{children:\"Replacing outdated methodology\"}),/*#__PURE__*/e(\"p\",{children:\"Virtually all automated translation follows the same logic, executed step by step:\"}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Separation of written content into smaller pieces\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Translation and validation of individual lexical components bit by bit\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Subsequent reassembling of the original message\"})})]}),/*#__PURE__*/e(\"img\",{alt:\"Traditional Pipeline\",className:\"framer-image\",height:\"517\",src:\"https://framerusercontent.com/images/0v2pAZLlwvvolOmTPuQ35mD0Kg.jpeg\",srcSet:\"https://framerusercontent.com/images/0v2pAZLlwvvolOmTPuQ35mD0Kg.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/0v2pAZLlwvvolOmTPuQ35mD0Kg.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/0v2pAZLlwvvolOmTPuQ35mD0Kg.jpeg 1840w\",style:{aspectRatio:\"1840 / 1035\"},width:\"920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Traditional Pipeline\"})}),/*#__PURE__*/e(\"p\",{children:\"All attempts at improving translation quality have so far been along the same line. Originally, Toloka was no different and followed the same tried-and-true pattern, which entailed going back and forth between translating and validating until an adequate result was achieved.\"}),/*#__PURE__*/e(\"p\",{children:'The problem is that this method is quite inefficient. It relies heavily on going through multiple stages (or \"projects\") and therefore requires a lot of time to execute. This issue of impracticality becomes even more pronounced when many projects are run in parallel. From a business perspective, this sort of setup is far from ideal.'}),/*#__PURE__*/e(\"h2\",{children:\"New translation algorithm\"}),/*#__PURE__*/e(\"p\",{children:\"Toloka programmers, led by Andrey Olkhovik, decided to skip the validation stage altogether, at least in the traditional form known to most in the auto-translation field.\"}),/*#__PURE__*/e(\"img\",{alt:\"New Experimental Approach\",className:\"framer-image\",height:\"517\",src:\"https://framerusercontent.com/images/oCh167swcgZyXAyF4gWjbvFXO4.jpeg\",srcSet:\"https://framerusercontent.com/images/oCh167swcgZyXAyF4gWjbvFXO4.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/oCh167swcgZyXAyF4gWjbvFXO4.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/oCh167swcgZyXAyF4gWjbvFXO4.jpeg 1840w\",style:{aspectRatio:\"1840 / 1035\"},width:\"920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"New Experimental Approach\"})}),/*#__PURE__*/e(\"p\",{children:'Now, Tolokers could choose one of the answers provided in a multiple-choice box containing up to 4 options, or tick \"none of the options fits\" and offer their own answer.'}),/*#__PURE__*/e(\"img\",{alt:\"Task example\",className:\"framer-image\",height:\"517\",src:\"https://framerusercontent.com/images/wmaSwHodlrGihkS7eKzphWVhO6E.jpeg\",srcSet:\"https://framerusercontent.com/images/wmaSwHodlrGihkS7eKzphWVhO6E.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/wmaSwHodlrGihkS7eKzphWVhO6E.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/wmaSwHodlrGihkS7eKzphWVhO6E.jpeg 1840w\",style:{aspectRatio:\"1840 / 1035\"},width:\"920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Task example\"})}),/*#__PURE__*/e(\"p\",{children:'As a result, the new versions provided by Tolokers became fixed options in a multiple-choice box for other contributors to choose from, and the whole \"choose the right option\" process started over.'}),/*#__PURE__*/t(\"p\",{children:['Different users either chose those answers and hence verified them, or offered their own, newer versions. The cycle continued until there were no other \"better\" answers to offer, and the best answers had been confirmed by most Tolokers.',/*#__PURE__*/e(\"br\",{}),/*#__PURE__*/e(\"br\",{className:\"trailing-break\"})]}),/*#__PURE__*/t(\"h2\",{children:[/*#__PURE__*/e(\"br\",{}),\"Results\"]}),/*#__PURE__*/e(\"p\",{children:\"Although this crowdsourcing-based method requires a much more careful selection of Tolokers, it provides a significant improvement in translation accuracy and offers the following advantages: 1 task \u2013 1 project. Both the metrics and the budget are focused, not scattered across numerous Toloka projects Less error-prone automatization. With reduced automatization, the potential for mistranslations also decreases Quick start. Less preparation is needed to begin the whole process Smaller budget. Not 2 projects for $0.01 + 50% commission, but 1 project for $0.02 + 25% commission\"})]});export const richText8=/*#__PURE__*/t(n.Fragment,{children:[/*#__PURE__*/t(\"p\",{children:[\"When a Japanese startup approached Toloka's partner \",/*#__PURE__*/e(a,{href:\"https://www.linkedin.com/in/kucev/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Roman Kucev\"})}),\" with 34,000 images from various TV shows and a seemingly daunting task of labeling human faces in every one of those images, they asked for 3 things: we want it done well, we want it done fast, and we want it done cheap... To the clients' delight, 3 weeks later, the task was completed at a fraction of the expected cost. \"]}),/*#__PURE__*/e(\"h2\",{children:\"Why crowdsourcing?\"}),/*#__PURE__*/e(\"p\",{children:\"Roman admits that even three years ago this task would have been tackled differently \u2014 without crowdsourcing \u2014 and it would have cost the clients 2.5 times the amount. Being a former employee of Prisma, Roman explains that other methods such as Computer Vision Annotation Tool (CVAT), though open-source and free, require a dedicated team of trained developers to run. Teams like that often aren't available. And their services are expensive.\"}),/*#__PURE__*/t(\"p\",{children:[\"Crowdsourcing has been a complete game changer that today allows companies to recruit talent across the board without needlessly paying through the roof. Instead of having a small team of highly qualified and often overpriced specialists do all of the work, \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/knowledgebase/crowdsourcing-concepts/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"crowdsourcing\"})}),\" allows for an infinitely larger pool of non-expert users, each one contributing a relatively small amount.\"]}),/*#__PURE__*/e(\"img\",{alt:\"Prices\",className:\"framer-image\",height:\"399\",src:\"https://framerusercontent.com/images/NvAPG1JP1QHiw11RMJAHjmfv84.jpeg\",srcSet:\"https://framerusercontent.com/images/NvAPG1JP1QHiw11RMJAHjmfv84.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/NvAPG1JP1QHiw11RMJAHjmfv84.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/NvAPG1JP1QHiw11RMJAHjmfv84.jpeg 1840w\",style:{aspectRatio:\"1840 / 798\"},width:\"920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Prices\"})}),/*#__PURE__*/e(\"h2\",{children:\"The task at hand and its challenges\"}),/*#__PURE__*/e(\"p\",{children:\"Since none of the content provided by the startup contained any personal data, crowdsourcing was a no-brainer. It was the only cost-effective way to go about the task of labeling tens of thousands of faces without having to hire software-specific experts. Be that as it may, the task still wasn't without its challenges.\"}),/*#__PURE__*/e(\"p\",{children:\"First, there was a bit of a disagreement as to what should be considered a human face. This may sound absurd at first, but it turned out that among the many images taken from a multitude of Japanese programs, there were not only those of men and women, but also anime characters, various drawings, human-like computer generated imagery, and humanoid androids. Eventually, it was decided that all but the animated characters and drawings were to be treated as human faces. \"}),/*#__PURE__*/e(\"img\",{alt:\"Face types\",className:\"framer-image\",height:\"767\",src:\"https://framerusercontent.com/images/Qh5YL9yuyzAlRTKmM08ideQ9n0.webp\",srcSet:\"https://framerusercontent.com/images/Qh5YL9yuyzAlRTKmM08ideQ9n0.webp?scale-down-to=512 512w,https://framerusercontent.com/images/Qh5YL9yuyzAlRTKmM08ideQ9n0.webp?scale-down-to=1024 1024w,https://framerusercontent.com/images/Qh5YL9yuyzAlRTKmM08ideQ9n0.webp?scale-down-to=2048 2048w,https://framerusercontent.com/images/Qh5YL9yuyzAlRTKmM08ideQ9n0.webp 2560w\",style:{aspectRatio:\"2560 / 1535\"},width:\"1280\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Face types\"})}),/*#__PURE__*/e(\"p\",{children:\"The next challenge was identifying different levels of blur and shakiness, different degrees of occlusion, and different poses \u2014 with follow-up instructions for Tolokers, which was key to accurate labeling.\"}),/*#__PURE__*/e(\"img\",{alt:\"Image parameters\",className:\"framer-image\",height:\"383\",src:\"https://framerusercontent.com/images/8OgWlm1Sy8uvTEKNYXYJAG7zHs.jpeg\",srcSet:\"https://framerusercontent.com/images/8OgWlm1Sy8uvTEKNYXYJAG7zHs.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/8OgWlm1Sy8uvTEKNYXYJAG7zHs.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/8OgWlm1Sy8uvTEKNYXYJAG7zHs.jpeg 1840w\",style:{aspectRatio:\"1840 / 766\"},width:\"920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Image parameters\"})}),/*#__PURE__*/e(\"p\",{children:\"Three colors were used (green, blue, and red), each one indicating a different rate of visibility.\"}),/*#__PURE__*/e(\"img\",{alt:\"Rate of visibility\",className:\"framer-image\",height:\"670\",src:\"https://framerusercontent.com/images/yE1oaxFYACf4HyY7c7ClR8bSTLY.jpeg\",srcSet:\"https://framerusercontent.com/images/yE1oaxFYACf4HyY7c7ClR8bSTLY.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/yE1oaxFYACf4HyY7c7ClR8bSTLY.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/yE1oaxFYACf4HyY7c7ClR8bSTLY.jpeg 1840w\",style:{aspectRatio:\"1840 / 1340\"},width:\"920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Rate of visibility\"})}),/*#__PURE__*/e(\"h2\",{children:\"Solution\"}),/*#__PURE__*/e(\"p\",{children:\"Every image could contain any number of faces, from zero to fifty. As a result, it was important to set different pay rates for processing images of varying complexity, and task-train all of the contributors. It was also necessary to assign a handful of moderators for quality control. The task was eventually solved in three stages:\"}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Introduction. Before starting on the task, every interested Toloker watched video instructions and then labeled 3-4 images as a test. If they did a good job, they moved on to the actual task.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Learning and labeling. Each stage of the task required a higher level of labeling skills: the Tolokers started out with images that contained only one face and gradually moved all the way up to 4+ faces. With this smooth learning curve, the Tolokers were more likely to deliver high quality on the more complex images. Each image took them around 7-8 minutes to label.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Quality control. A moderator, who was a more experienced user, subsequently checked whether each image was labeled correctly, which took an additional 10-15 seconds per image. Each moderator oversaw a team of 30-40 Tolokers on average.\"})})]}),/*#__PURE__*/e(\"img\",{alt:\"Solution\",className:\"framer-image\",height:\"385\",src:\"https://framerusercontent.com/images/zsoNjNSklsXVgZVLTESVT7dHvw.jpeg\",srcSet:\"https://framerusercontent.com/images/zsoNjNSklsXVgZVLTESVT7dHvw.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/zsoNjNSklsXVgZVLTESVT7dHvw.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/zsoNjNSklsXVgZVLTESVT7dHvw.jpeg 1840w\",style:{aspectRatio:\"1840 / 771\"},width:\"920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Solution\"})}),/*#__PURE__*/e(\"h2\",{children:\"Results\"}),/*#__PURE__*/e(\"p\",{children:\"65,000 faces were labeled over a period of 3 weeks with the cost of approximately $0.015 per face. The cost is estimated to be 250% lower than any other non-crowdsourcing solution currently available on the market while the quality never fell below market average throughout. \"}),/*#__PURE__*/e(\"img\",{alt:\"Results\",className:\"framer-image\",height:\"509\",src:\"https://framerusercontent.com/images/wR1Ov8WlPSSIGv5S3lQuQxp8x6s.jpeg\",srcSet:\"https://framerusercontent.com/images/wR1Ov8WlPSSIGv5S3lQuQxp8x6s.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/wR1Ov8WlPSSIGv5S3lQuQxp8x6s.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/wR1Ov8WlPSSIGv5S3lQuQxp8x6s.jpeg 1840w\",style:{aspectRatio:\"1840 / 1018\"},width:\"920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Results\"})})]});export const richText9=/*#__PURE__*/t(n.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"Tolokers hail from all over the globe, from a multitude of age groups and occupations. We already introduced you to some of them a while back. This time, we have three personal stories to share from Latin America, Asia, and Africa. \"}),/*#__PURE__*/t(\"p\",{children:[\"We love learning about ambitious Tolokers and finding out why individuals join Toloka. As it turns out, some use Toloka as a stepping stone to their professional goals, like Damaris, a smiley 20-something South American who splits her time between earning a Master's degree and managing a store. Some have decided to get on the \\\"crowdsource express\\\" after seeing its full potential in the context of the AI boom, like Nguyen Diep, an active businessman from Vietnam. Yet others are earning money to help with expenses while they're getting a degree, like Alfred from Nigeria. Our three contributors agreed to sit down for a talk to explain what attracted them to \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/knowledgebase/crowdsourcing-concepts/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"crowdsourcing\"})}),\" in general and Toloka in particular, and what exactly they get out of being Tolokers.\"]}),/*#__PURE__*/e(\"img\",{alt:\"Three Muskateers\",className:\"framer-image\",height:\"288\",src:\"https://framerusercontent.com/images/cIKBsBM6Rusmx3KgIoKx7sfBoo.jpeg\",srcSet:\"https://framerusercontent.com/images/cIKBsBM6Rusmx3KgIoKx7sfBoo.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/cIKBsBM6Rusmx3KgIoKx7sfBoo.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/cIKBsBM6Rusmx3KgIoKx7sfBoo.jpeg 1840w\",style:{aspectRatio:\"1840 / 576\"},width:\"920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Three Muskateers\"})}),/*#__PURE__*/e(\"h2\",{children:\"Please tell us about yourself\"}),/*#__PURE__*/e(\"p\",{children:\"Damaris: My name is Damaris Elizabeth Briones, I'm 24 years old, and I come from Argentina. I'm in the middle of earning a university degree, and I also manage a computer repair shop \u2014 in addition to being a dedicated Toloker, naturally.\"}),/*#__PURE__*/e(\"p\",{children:\"Nguyen Diep: My name is Nguyen Quang Diep, I'm 40 years old, and I'm from Vietnam. I've been focused on business for quite a while. I'm basically attracted to any industry that can get me cash. These days it's IT.\"}),/*#__PURE__*/e(\"p\",{children:\"Alfred: My name is Alfred Mathias, I'm 19 years old, and I'm from Abuja, the capital city of Nigeria, where I've recently started college. I'm a big tech enthusiast, and for the past several months I've also been a Toloker.\"}),/*#__PURE__*/e(\"h2\",{children:\"How did you find out about Toloka and why did you decide to join?\"}),/*#__PURE__*/e(\"p\",{children:\"Damaris: I was looking for a part-time job to moderate online content or do something related to website management and telecommunications, because this is what I want to pursue professionally in the future. When I found Toloka, I knew that it was the perfect fit because of Toloka's involvement in content research and data labeling. I immediately discovered that I liked the topics of the tasks, how easy the tasks are, and how quickly they can be completed. That and how the whole platform is organized.\"}),/*#__PURE__*/e(\"p\",{children:\"Nguyen Diep: Before Toloka, I did some crowdsourcing and web surveys on other platforms. What attracted me to Toloka is its reliability, and also the fact that the tasks are thought-provoking and even educational. I was genuinely surprised that what Toloka was offering was so different, particularly when compared to what else is available out there, which is often extremely boring, to be honest.\"}),/*#__PURE__*/e(\"p\",{children:\"Alfred: A good friend of mine told me about Toloka. What caught my attention is the fact that \u2014 unlike many other crowdsourcing platforms \u2014 Toloka doesn't make you put up any money upfront to join. And what's more, the service is totally transparent, without any hidden fees or other unpleasant surprises: you get paid precisely for the work that you've done, and always on time.\"}),/*#__PURE__*/e(\"h2\",{children:\"When did you start doing your first tasks?\"}),/*#__PURE__*/e(\"p\",{children:\"Damaris: In the beginning of last year. Nguyen Diep: I've been involved in crowdsourcing for over 6 years now, and the last 3 years I've devoted most of my time to Toloka. Alfred: I joined Toloka in December of last year and haven't looked back since. So far, so good!\"}),/*#__PURE__*/e(\"h2\",{children:\"What types of tasks do you prefer and why?\"}),/*#__PURE__*/t(\"p\",{children:[\"Damaris: Obviously, I prefer the tasks that pay better. But aside from price, I really like to classify pictures, posts on social media, and articles. I also enjoy classifying web searches, in both English and Spanish. Nguyen Diep: I do many different \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/data-labeling-platform/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"data labeling\"})}),\" tasks for Toloka, but I particularly enjoy hands-on field tasks and surveys, because they give me a reason to go somewhere. I like to be able to explore a specific industry or work environment, be it a supermarket, taxi company, or delivery service. On the other hand, I like to be able to express my opinion and provide useful feedback. Alfred: There are a number of them, but I prefer mainly voice annotation tasks. To me, those are quite engaging, which makes the whole process easy and enjoyable. I don't mind them at all. I'd also like to do more website optimization and robot voice tasks in the future. The latter, in particular, pushes the envelope of Machine Learning and Artificial Intelligence development. I find it very exciting.\"]}),/*#__PURE__*/e(\"h2\",{children:\"How often do you work and how do you spend the money you earn with Toloka?\"}),/*#__PURE__*/e(\"p\",{children:\"Damaris: I try to put in at least a couple of hours daily. Provided I keep up the pace, I end up with quite a bit in my pocket at the end of every month, which I use in any way I like, really. Nguyen Diep: I prefer to work at a leisurely pace, approximately 60 to 90 minutes a day. Sometimes more if I have more free time on my hands, or I feel like making a bit more money. I spend what I earn mainly on gas and medication for my parents, as well as groceries, coffee, and clean water. Alfred: This is another beautiful thing: I don't spend more than 1 hour every day, sometimes even less. Given the benefits I get out of it, I think this isn't a great deal of effort on my part, as it doesn't eat up a lot of my free time. How much I make varies depending on how busy I am at college and how many tasks I can handle. And also the task price, of course. The money I earn is a nice addition towards my living expenses: I can easily cover some of my basic needs with it and also have enough left over for my education (textbooks, stationery, administrative fees, etc).\"}),/*#__PURE__*/e(\"h2\",{children:\"What do you like about Toloka (other than making money)?\"}),/*#__PURE__*/e(\"p\",{children:\"Damaris: I like that it fits in nicely with my lifestyle and goes hand in hand with what I'm already doing. It doesn't take away from my job, and it doesn't interfere with my university studies \u2014 on the contrary, it teaches me relevant skills, and all in all, keeps me on my toes professionally. Nguyen Diep: In addition to having a supplementary income, I really enjoy getting to express my views on the world around us, which I often don't get to do in my main business. I also learn useful things every day, an opportunity that would have been lost otherwise. And I appreciate that taking out my earnings is super easy and straightforward with Toloka. Alfred: I really like that Toloka has no withdrawal limit, so it means I'm free to do whatever I like with the money I make, which is a big deal as far as I'm concerned. Also, it's nice to feel that I'm contributing to a greater goal, and my efforts have a real purpose in the grand scheme of things.\"}),/*#__PURE__*/e(\"h2\",{children:\"What are the advantages of being a data labeler?\"}),/*#__PURE__*/e(\"p\",{children:\"Damaris: I think it's fun. It's even more fun knowing that you get paid for it. Also, it's a great way to stay up to date and even ahead of the curve in many regards as far as the latest technology goes. And, of course, it's very fulfilling to know that you're helping people and improving future user experiences on the web and with electronics in general. Nguyen Diep: Being able to practice memory skills and develop mental agility, training listening and spelling skills, and also acquiring valuable tech knowledge in the process. Alfred: For me, it's mainly about convenience. I can do tasks whenever and as much as I want to. I can choose whatever tasks I prefer. And, importantly, I can access tasks anytime from anywhere.\"}),/*#__PURE__*/e(\"h2\",{children:\"What would you tell someone who is thinking about trying Toloka?\"}),/*#__PURE__*/e(\"p\",{children:\"Damaris: Toloka is a great way to learn about the Internet and have an easy and flexible part-time job that brings in regular cash. If you like working with websites, no matter your preferred device, you can find something suitable with Toloka. It's also a great way to rapidly acquire new skills and increase your own value on the market. Nguyen Diep: I'd say definitely go ahead and try, but always remain diligent and honest. I never slack off and always complete all tasks on my own, to the best of my abilities, and that gives me a lot of satisfaction. Alfred: Give it a shot! Toloka is not just an empty promise. It is a truly dependable crowdsourcing service, and if you put in some time, you will get paid fair and square for your efforts, no questions asked. Which tasks you tackle and how much you earn is entirely up to you.\\xa0\"})]});export const richText10=/*#__PURE__*/t(n.Fragment,{children:[/*#__PURE__*/t(\"p\",{children:[\"Humans make tons of waste, especially in cities. Urban waste contributes heavily to greenhouse gas emissions, impacts groundwater quality, and attracts rats and other disease-carrying animals. Trash generation is growing rapidly but so is the number of initiatives aimed at proper waste management. \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/knowledgebase/crowdsourcing-concepts/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Crowdsourcing\"})}),\" platforms like Toloka can help with that. Here's how. \"]}),/*#__PURE__*/e(\"h2\",{children:\"Rating trash container sites with Toloka\"}),/*#__PURE__*/e(\"p\",{children:\"Toloka recently checked public trash containers to rate the cleanest Russian cities. Tolokers identified overflowing dumpsters, checked garbage collection schedules, and assessed the overall state of trash container sites. The contributors were ordinary citizens \u2013 people who pass trash containers every day on their way to work or simply use them to dispose of their garbage.\"}),/*#__PURE__*/e(\"p\",{children:\"Their task was to take a picture of a trash container site and answer a few questions about it in the Toloka app. This information was then added to the site's description in Yandex.Maps.\"}),/*#__PURE__*/e(\"h2\",{children:\"How it's done\"}),/*#__PURE__*/e(\"p\",{children:\"To assess and rate a public trash collection site, contributors followed these simple steps:\"}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:'Open the Toloka mobile app and go to the \"Check a waste container site\" task.'})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Find an unrated container on the map and go to that location.\"})})]}),/*#__PURE__*/e(\"img\",{alt:\"Find a container\",className:\"framer-image\",height:\"1080\",src:\"https://framerusercontent.com/images/5c4iGcGqWeeRtzw0LxQ6oRRlsCc.jpeg\",srcSet:\"https://framerusercontent.com/images/5c4iGcGqWeeRtzw0LxQ6oRRlsCc.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/5c4iGcGqWeeRtzw0LxQ6oRRlsCc.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/5c4iGcGqWeeRtzw0LxQ6oRRlsCc.jpeg?scale-down-to=2048 2048w,https://framerusercontent.com/images/5c4iGcGqWeeRtzw0LxQ6oRRlsCc.jpeg 3840w\",style:{aspectRatio:\"3840 / 2160\"},width:\"1920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Find a container\"})}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Once there, take 3 or 4 pictures of the container and upload them to Toloka.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Answer the questions in the task: Is the area paved? Is there a roof or any sort of awning over the container? Is there any trash lying outside the container? What is the garbage collection schedule? Are there separate containers for recycling?\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Submit the input.\"})})]}),/*#__PURE__*/e(\"img\",{alt:\"Take pictures, answer the questions\",className:\"framer-image\",height:\"1080\",src:\"https://framerusercontent.com/images/k2hMtNcVSFDsJlp3aGLO3eII.jpeg\",srcSet:\"https://framerusercontent.com/images/k2hMtNcVSFDsJlp3aGLO3eII.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/k2hMtNcVSFDsJlp3aGLO3eII.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/k2hMtNcVSFDsJlp3aGLO3eII.jpeg?scale-down-to=2048 2048w,https://framerusercontent.com/images/k2hMtNcVSFDsJlp3aGLO3eII.jpeg 3840w\",style:{aspectRatio:\"3840 / 2160\"},width:\"1920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Take pictures, answer the questions\"})}),/*#__PURE__*/e(\"h2\",{children:\"Results\"}),/*#__PURE__*/e(\"p\",{children:\"In three months, Tolokers managed to rate a whopping 74,686 trash container sites in 58 regions of Russia. Their efforts resulted in a rating of the cleanest and dirtiest cities and helped raise awareness about problems with garbage disposal in some regions.\"}),/*#__PURE__*/e(\"p\",{children:\"For example, only 11% of the containers inspected in North Ossetia were clean, so now the regional government reviews the rating every three months to identify problematic sites and adjust the work of the waste management companies that oversee them.\"}),/*#__PURE__*/e(\"p\",{children:\"Top 3 regions with the best waste container site management:\"}),/*#__PURE__*/t(\"ol\",{children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Moscow\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Yaroslavl Oblast\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Moscow Oblast\"})})]}),/*#__PURE__*/e(\"p\",{children:\"Improved waste container site management since the project's launch:\"}),/*#__PURE__*/t(\"ol\",{children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Sakhalin Oblast\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Tyumen Oblast\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Tula Oblast \"})})]})]});export const richText11=/*#__PURE__*/t(n.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"Tolokers are people from all over the world and all walks of life \u2014 from young people just starting out to professionals with extensive experience. Our users have a fascinating mix of backgrounds, languages, occupations, hobbies and interests, but there is one thing they all share: Toloka fits into their lifestyle. In this post we'll introduce you to three Tolokers from India, Brazil, and Russia. \"}),/*#__PURE__*/e(\"h2\",{children:\"Pranjal: engineer and YouTube blogger\"}),/*#__PURE__*/e(\"p\",{children:\"You don't need a degree to be successful in Toloka, but many users do have tech degrees \u2014 like Pranjal, a Toloker from India. He worked as an engineer at a telecommunications company, but lost his job due to layoffs. We talked to Pranjal about how Toloka helps him earn extra income and develop skills in the field of artificial intelligence.\"}),/*#__PURE__*/e(\"img\",{alt:\"Pranjal: engineer and YouTube blogger\",className:\"framer-image\",height:\"1706\",src:\"https://framerusercontent.com/images/3W4xTZRVJGuibfjApMh79Tu6M.webp\",srcSet:\"https://framerusercontent.com/images/3W4xTZRVJGuibfjApMh79Tu6M.webp?scale-down-to=1024 768w,https://framerusercontent.com/images/3W4xTZRVJGuibfjApMh79Tu6M.webp?scale-down-to=2048 1536w,https://framerusercontent.com/images/3W4xTZRVJGuibfjApMh79Tu6M.webp 2560w\",style:{aspectRatio:\"2560 / 3413\"},width:\"1280\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Pranjal\"})}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"Tell us about yourself\"}),\" I'm 26 years old. I have a Bachelor of Science in Engineering with a focus on instrumentation and applied electronics. I live in Gauhati (Assam, India). I started working for a telecommunications company at the airport, but I was let go because of the pandemic. I'm currently looking for a permanent job.\"]}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"What are your hobbies?\"}),\" I'm interested in everything related to the internet, technology, and online games. I also love movies and I have a YouTube channel where I talk about the newest gadgets.\"]}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"When did you start doing your first tasks in Toloka?\"}),\" In the fall of 2020, I left my job and started looking for a way to make money online. In November, I saw an ad on Facebook, installed the app, and I've been doing tasks ever since. On average, I spend 3-5 hours a day in Toloka.\"]}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"What are your favorite tasks or some unusual tasks you've done?\"}),\" I prefer to take tasks from requesters who pay instantly. Sometimes it takes 7 to 15 days for tasks to get checked and accepted. I was a little skeptical about tasks where you make a video of your face, but then I read the instructions carefully and realized that this is important for training artificial intelligence.\"]}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"How much do you earn and what do you spend it on?\"}),\" I've been doing tasks in Toloka every day since November 2020. I'm earning about $65 a month on average. I mostly spend it on traveling, but I also buy little things for the house and gas for my motorcycle.\"]}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"What are the advantages of being a data labeler?\"}),\" You don't need a special degree or experience to learn how artificial intelligence and machine learning work and try out this field. Another bonus is that you can do tasks from anywhere in the world, whenever you feel like it. After you've completed a few tasks, you can consider yourself an expert \u2014 data labelers learn as they go. The more experience you get, the more knowledge you have.\"]}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"What other tasks would you like to do?\"}),\" I'm interested in tasks related to typing and transcribing text, as well as evaluating \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/search-relevance/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"search relevance\"})}),\", when you need to check how well site pages match the user's search query. I want to get more tasks with higher rewards.\"]}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"What do you like about Toloka (other than making money)?\"}),\" I like that the interface is minimalistic and intuitive. No ads \u2014 and the tasks are interesting. In my opinion, this makes Toloka one of the best platforms for earning money online.\"]}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"Do you have any advice for people who are thinking about signing up for Toloka?\"}),' Go through training, read the instructions carefully and practice as many skills as you can in different tasks. Take the time to do the training tasks. You\\'ll need the skills when you start doing the \"real\" tasks. Good luck!']}),/*#__PURE__*/e(\"h2\",{children:\"Andr\\xe9: Surfer and dad\"}),/*#__PURE__*/e(\"p\",{children:\"What do surfing, house renovations and Toloka have in common? Andr\\xe9 from Brazil says they go hand in hand as a way to balance his free time.\"}),/*#__PURE__*/e(\"img\",{alt:\"Andr\\xe9: Surfer and dad\",className:\"framer-image\",height:\"1706\",src:\"https://framerusercontent.com/images/PGNWYxEgXNDImogP6QF3X0IfHBw.webp\",srcSet:\"https://framerusercontent.com/images/PGNWYxEgXNDImogP6QF3X0IfHBw.webp?scale-down-to=1024 768w,https://framerusercontent.com/images/PGNWYxEgXNDImogP6QF3X0IfHBw.webp?scale-down-to=2048 1536w,https://framerusercontent.com/images/PGNWYxEgXNDImogP6QF3X0IfHBw.webp 2560w\",style:{aspectRatio:\"2560 / 3413\"},width:\"1280\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Andr\\xe9\"})}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"Tell us about yourself\"}),\" My name is Andr\\xe9 Lerman. I'm from Brazil and I'm a medical radiology technician. I am married and I have a stepson \u2013 he's a great kid and we do a lot of cycling and skateboarding together. I absolutely love surfing. Brazil has some gorgeous beaches and surfing locations, and I do my best to take full advantage of that. Another thing I really enjoy is taking care of the house, doing small renovations and looking after my plants. I am also a performer at Toloka.\"]}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"How did you join Toloka?\"}),\" I was looking for a way to make some extra money and I saw that Toloka is a serious and committed platform that treats its collaborators fairly and has a qualified support team. I decided to try it out, and I got into it really quickly. I spend more time in Toloka now than I did in the beginning: so far, my personal record is 2 hours straight. I manage to squeeze it into my routine every day.\"]}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"What are your favorite tasks or some unusual tasks you've done?\"}),\" I like all the tasks in Toloka but there are some I find particularly interesting. My favorites are outlining objects in photos and the tasks where you need to count animals. I also enjoy sending short videos and I really want to do more of that. The most unusual task I've had so far was mapping categories. I hadn't done it before, but I really liked it and I hope to get better at it. Going forward, I want to up my game and get good enough to be offered more paid video tasks.\"]}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"What do you like about data labeling?\"}),\" There are a few reasons why \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/data-labeling-platform/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"data labeling\"})}),\" is a great field. For me, the main advantage is that I can do it anytime, anywhere. Nobody cares if I'm sitting in bed, you know? I think that's pretty awesome.\"]}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"Do you have any advice for people who are thinking about signing up for Toloka?\"}),\" If you're thinking of joining Toloka, go for it! There's a good chance you'll like it, just like I did. With Toloka, you choose when and where to do stuff. You're basically your own boss. Plus, it's a serious company that pays you real money \u2013 no delays, no bureaucracy, no inconvenience.\"]}),/*#__PURE__*/e(\"h2\",{children:\"Nadia: dog lover and volunteer\"}),/*#__PURE__*/e(\"p\",{children:\"Nadia Teterina is an animal shelter volunteer who has seamlessly integrated Toloka into her everyday life: she completes tasks on her way to the store and when she's walking her dogs.\"}),/*#__PURE__*/e(\"img\",{alt:\"Nadia: dog lover and volunteer\",className:\"framer-image\",height:\"1706\",src:\"https://framerusercontent.com/images/EfbljydfrKH0LnLk3egydI3Ho.webp\",srcSet:\"https://framerusercontent.com/images/EfbljydfrKH0LnLk3egydI3Ho.webp?scale-down-to=1024 768w,https://framerusercontent.com/images/EfbljydfrKH0LnLk3egydI3Ho.webp?scale-down-to=2048 1536w,https://framerusercontent.com/images/EfbljydfrKH0LnLk3egydI3Ho.webp 2560w\",style:{aspectRatio:\"2560 / 3413\"},width:\"1280\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Nadia\"})}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"What do you do for a living?\"}),\" I design websites with Tilda and do advertising for a print magazine. I've been working remotely for 10 years, so last year's lockdown wasn't a big change for me. My work-from-home arrangement has allowed me to take care of my dogs, one of whom suffers from spinal damage.\"]}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"How did you join Toloka?\"}),\" I spent a great deal of last summer playing games on my phone. I got sick of it by early fall and deleted everything except sudoku. I'd seen Toloka ads a few times and I thought, \\\"Hey, let's see if I can make some money off of it\\\". Toloka turned out to be a great fit for me: it's like playing a game \u2013 but you also get paid! I even quit sudoku. I do tasks in Toloka after waking up and before going to bed. Sometimes I manage to squeeze in a few tasks while I'm walking my dogs. All in all, I spend around 2 hours a day in Toloka.\"]}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"What are your favorite tasks?\"}),\" I like to check the relevance of search query responses and classify messages posted on social media. I took a speed-reading course a few years back, and now it really helps me complete these tasks faster.\"]}),/*#__PURE__*/e(\"p\",{children:\"I also enjoy taking pictures. Every time I'm about to go to the post office or to get groceries, I check Toloka for field tasks to complete on my way, like taking a photo of a building or a business's entrance. I think it's a great way to get some extra exercise and earn a few dollars at the same time.\"}),/*#__PURE__*/e(\"p\",{children:\"Side-by-side comparison of images is fun, but I found it a tiny bit disconcerting. Picture it: you're sitting there, comparing images of completely innocent things like clothes or footwear. Then suddenly, out of the blue, comes a picture of somebody's privates. It happened pretty often. I can't even begin to describe the things I've seen! When a friend of mine joined Toloka and started doing these tasks, we talked about it a lot. The good news is that you don't have to do it if you feel uncomfortable: you can disable tasks with adult content in settings so that you don't have to deal with anything inappropriate.\"}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"How do you spend what you earn in Toloka?\"}),\" I spend my earnings on phone bills, small gifts for myself, and charity. I am an animal shelter volunteer, and it's one of the things that drives me. Most of the money I earn I donate to animal shelters, so the more money I make in Toloka, the more animals I get to help.\"]}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"What do you like about data labeling?\"}),\" I really enjoy the feeling of being involved in a meaningful project. When I was little, I watched sci-fi movies and couldn't imagine myself living in a future like today or that I'd help train artificial intelligence.\"]}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"Do you have any advice for people who are thinking about signing up for Toloka?\"}),\" If you're ready to set aside at least a couple of hours a day, and if you're patient and able to concentrate, you absolutely can earn some extra cash with Toloka. There's no need to quit your current job \u2013 you can complete tasks in your spare time. You choose your own schedule and task complexity, and that's really convenient.\"]})]});export const richText12=/*#__PURE__*/t(n.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"Toloka went through some massive changes this week. We launched a full-scale rebrand and revealed a new logo and brand identity. Our sites and products now have a more refined, crisp look that we hope you're enjoying as much as we are. We put our hearts and souls into this extensive makeover, and we're excited to finally share it with you. Keep reading to find out how we went from our old daisy logo to a dandelion that's also a bouquet of popsicles.  \"}),/*#__PURE__*/e(\"h2\",{children:\"Why we decided to rebrand\"}),/*#__PURE__*/e(\"p\",{children:\"Since its launch in 2014, Toloka has grown a great deal. For us, it meant working with new clients, meeting new users, and taking a completely different approach to how we present ourselves.\"}),/*#__PURE__*/t(\"p\",{children:[\"More importantly, we interact with two very different parts of the \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/knowledgebase/crowdsourcing-concepts/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"crowdsourcing\"})}),\" world. One is all about the requester and their \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/data-labeling-platform/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"data labeling\"})}),\" needs, and the other is about the performer, who wants to contribute to something big and earn some extra cash. The former focuses on technology, innovation, and science, while the latter is centered around things like collaboration, opportunity, and freedom.\"]}),/*#__PURE__*/e(\"img\",{alt:\"Brand attributes\",className:\"framer-image\",height:\"1035\",src:\"https://framerusercontent.com/images/g7Rvw9q4OGYJvuIWII8uFP5oZrc.jpeg\",srcSet:\"https://framerusercontent.com/images/g7Rvw9q4OGYJvuIWII8uFP5oZrc.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/g7Rvw9q4OGYJvuIWII8uFP5oZrc.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/g7Rvw9q4OGYJvuIWII8uFP5oZrc.jpeg?scale-down-to=2048 2048w,https://framerusercontent.com/images/g7Rvw9q4OGYJvuIWII8uFP5oZrc.jpeg 3680w\",style:{aspectRatio:\"3680 / 2070\"},width:\"1840\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Brand attributes\"})}),/*#__PURE__*/e(\"p\",{children:\"We want both requesters and Tolokers to feel comfortable and make the most out of their experience with Toloka \u2013 but our old brand identity didn't quite represent the fundamental differences between these two groups. Plus, we knew about cases when new requesters tried registering as a performer by mistake, and vice versa. It was time to draw a more distinctive line between the two.\"}),/*#__PURE__*/e(\"p\",{children:\"To that end, we defined two main goals (and, honestly, they only seem mutually exclusive):\"}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Visually convey the individual attributes of each group and put an end to all the confusion.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Develop a shared visual concept for our products to show that both requesters and Tolokers are a part of the platform.\"})})]}),/*#__PURE__*/e(\"p\",{children:\"As hard as it was to reconcile the two, we managed to create two individual brand designs that obviously share the same story but tell it from different perspectives: a sleek and premium-feel sapphire blue on dark gray for the requesters, and a warmer, cozier salmon and light gray for the Tolokers.\"}),/*#__PURE__*/e(\"img\",{alt:\"Color Scheme\",className:\"framer-image\",height:\"1035\",src:\"https://framerusercontent.com/images/zL0pGd4V420FWJ4Pqo4kqKYMQ.jpeg\",srcSet:\"https://framerusercontent.com/images/zL0pGd4V420FWJ4Pqo4kqKYMQ.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/zL0pGd4V420FWJ4Pqo4kqKYMQ.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/zL0pGd4V420FWJ4Pqo4kqKYMQ.jpeg?scale-down-to=2048 2048w,https://framerusercontent.com/images/zL0pGd4V420FWJ4Pqo4kqKYMQ.jpeg 3680w\",style:{aspectRatio:\"3680 / 2070\"},width:\"1840\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Color Scheme\"})}),/*#__PURE__*/e(\"h2\",{children:\"The concept: why we chose fractals\"}),/*#__PURE__*/e(\"p\",{children:\"Our ideas for the brand concept were based on our associations with crowdsourcing. We talked about bounding box shapes and labeling points; noisy data and how it transitions into meaningful data (see those blurry circles that gradually come into focus?) and we talked about fractals.\"}),/*#__PURE__*/e(\"img\",{alt:\"Concepts\",className:\"framer-image\",height:\"1035\",src:\"https://framerusercontent.com/images/sbsxBdAzvuJIftseGUp6vCeyUd4.jpeg\",srcSet:\"https://framerusercontent.com/images/sbsxBdAzvuJIftseGUp6vCeyUd4.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/sbsxBdAzvuJIftseGUp6vCeyUd4.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/sbsxBdAzvuJIftseGUp6vCeyUd4.jpeg?scale-down-to=2048 2048w,https://framerusercontent.com/images/sbsxBdAzvuJIftseGUp6vCeyUd4.jpeg 3680w\",style:{aspectRatio:\"3680 / 2070\"},width:\"1840\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Concepts\"})}),/*#__PURE__*/e(\"p\",{children:\"Spoiler alert: fractals won.\"}),/*#__PURE__*/e(\"img\",{alt:\"Fractals\",className:\"framer-image\",height:\"1035\",src:\"https://framerusercontent.com/images/OyKz2u9SX1BpgIS6GQQTVLZ0BPk.jpeg\",srcSet:\"https://framerusercontent.com/images/OyKz2u9SX1BpgIS6GQQTVLZ0BPk.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/OyKz2u9SX1BpgIS6GQQTVLZ0BPk.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/OyKz2u9SX1BpgIS6GQQTVLZ0BPk.jpeg?scale-down-to=2048 2048w,https://framerusercontent.com/images/OyKz2u9SX1BpgIS6GQQTVLZ0BPk.jpeg 3680w\",style:{aspectRatio:\"3680 / 2070\"},width:\"1840\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Fractals\"})}),/*#__PURE__*/e(\"p\",{children:\"We think fractals are beautiful. They are present in the natural world in everything from tree leaves to the brilliant human mind, and in mathematics, which lies at the heart of all modern technology.\"}),/*#__PURE__*/e(\"img\",{alt:\"\",className:\"framer-image\",height:\"401\",src:\"https://framerusercontent.com/images/zm2cUzHZUpBUxpNDYSpv4O9DLc.jpeg\",srcSet:\"https://framerusercontent.com/images/zm2cUzHZUpBUxpNDYSpv4O9DLc.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/zm2cUzHZUpBUxpNDYSpv4O9DLc.jpeg 804w\",style:{aspectRatio:\"804 / 802\"},width:\"402\"}),/*#__PURE__*/e(\"p\",{children:\"What's even better is how fractals represent the very essence of crowdsourcing, where a multitude of basic forms, put together, shape something that's highly complex.\"}),/*#__PURE__*/e(\"img\",{alt:\"Fractals\",className:\"framer-image\",height:\"540\",src:\"https://framerusercontent.com/images/pwrHZgiSu6QVxLRvCxYNFAM7Weo.jpeg\",srcSet:\"https://framerusercontent.com/images/pwrHZgiSu6QVxLRvCxYNFAM7Weo.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/pwrHZgiSu6QVxLRvCxYNFAM7Weo.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/pwrHZgiSu6QVxLRvCxYNFAM7Weo.jpeg 1920w\",style:{aspectRatio:\"1920 / 1080\"},width:\"960\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Fractals\"})}),/*#__PURE__*/e(\"h2\",{children:\"The logo: a hard-won flower\"}),/*#__PURE__*/e(\"p\",{children:\"One of the toughest parts of our rebranding experience was choosing a new logo. The idea was to use fractals and try to somehow preserve the daisy that so many people on our team love. We looked at different flowers, but the first attempts at making a floral-based logo didn't work out: the icons weren't scalable, and some people had associations with propellers or even the swastika \u2013 which certainly wasn't what we were going for.\"}),/*#__PURE__*/e(\"img\",{alt:\"Logo concepts\",className:\"framer-image\",height:\"800\",src:\"https://framerusercontent.com/images/LA0QNZ2n8owkg3b3egpeghz0cA.jpeg\",srcSet:\"https://framerusercontent.com/images/LA0QNZ2n8owkg3b3egpeghz0cA.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/LA0QNZ2n8owkg3b3egpeghz0cA.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/LA0QNZ2n8owkg3b3egpeghz0cA.jpeg?scale-down-to=2048 2048w,https://framerusercontent.com/images/LA0QNZ2n8owkg3b3egpeghz0cA.jpeg 2845w\",style:{aspectRatio:\"2845 / 1600\"},width:\"1422\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Logo concepts\"})}),/*#__PURE__*/e(\"p\",{children:\"Deciding it was time to move on and forget about the daisy altogether, we changed tack and tried combining basic forms like circles and ovals.\"}),/*#__PURE__*/e(\"img\",{alt:\"Logo concepts\",className:\"framer-image\",height:\"1035\",src:\"https://framerusercontent.com/images/4RbKk4M9ELC7Urv8DzcVL7x4SyU.jpeg\",srcSet:\"https://framerusercontent.com/images/4RbKk4M9ELC7Urv8DzcVL7x4SyU.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/4RbKk4M9ELC7Urv8DzcVL7x4SyU.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/4RbKk4M9ELC7Urv8DzcVL7x4SyU.jpeg?scale-down-to=2048 2048w,https://framerusercontent.com/images/4RbKk4M9ELC7Urv8DzcVL7x4SyU.jpeg 3680w\",style:{aspectRatio:\"3680 / 2070\"},width:\"1840\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Logo concepts\"})}),/*#__PURE__*/e(\"p\",{children:\"But nothing seemed like a good fit: test results showed that our ideas were either very similar to existing logos, or evoked weird associations with sacral chakra symbols. So, we changed course once again and ended up with five finalists that both represented our floral preferences and carried a sharp, modern look.\"}),/*#__PURE__*/e(\"img\",{alt:\"Finalists\",className:\"framer-image\",height:\"1035\",src:\"https://framerusercontent.com/images/nUk3Hapfyu2zQBzhXmKoWo0Ww.jpeg\",srcSet:\"https://framerusercontent.com/images/nUk3Hapfyu2zQBzhXmKoWo0Ww.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/nUk3Hapfyu2zQBzhXmKoWo0Ww.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/nUk3Hapfyu2zQBzhXmKoWo0Ww.jpeg?scale-down-to=2048 2048w,https://framerusercontent.com/images/nUk3Hapfyu2zQBzhXmKoWo0Ww.jpeg 3680w\",style:{aspectRatio:\"3680 / 2070\"},width:\"1840\"}),/*#__PURE__*/e(\"p\",{children:\"Finalists\"}),/*#__PURE__*/e(\"p\",{children:\"Who do you think helped us pick the winner? Tolokers from around the globe, of course. We asked them to share their thoughts about each logo, and we got some great insights. A few of them really cracked us up, too. Take a look at these logos and check if you can see a death star, a toilet sign, prison bars, and Pacman.\"}),/*#__PURE__*/e(\"p\",{children:\"As amusing as it sounds, these aren't the associations we want people to have with Toloka. Without a second thought, we chose the logo that garnered the brightest, warmest and most positive associations, like ice cream, dandelions, and the golden ratio.\"}),/*#__PURE__*/e(\"img\",{alt:\"Our logo\",className:\"framer-image\",height:\"1035\",src:\"https://framerusercontent.com/images/i0M5xR1zuaMaSgHBawbS1TD9LHc.jpeg\",srcSet:\"https://framerusercontent.com/images/i0M5xR1zuaMaSgHBawbS1TD9LHc.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/i0M5xR1zuaMaSgHBawbS1TD9LHc.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/i0M5xR1zuaMaSgHBawbS1TD9LHc.jpeg?scale-down-to=2048 2048w,https://framerusercontent.com/images/i0M5xR1zuaMaSgHBawbS1TD9LHc.jpeg 3680w\",style:{aspectRatio:\"3680 / 2070\"},width:\"1840\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Our logo\"})}),/*#__PURE__*/e(\"p\",{children:\"Doesn't it feel like the perfect logo for Toloka? Sure, we're biased, but think about it:\"}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"It doesn't seem to evoke associations with already existing logos.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"It's a flower.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Even though it's not a fractal in the strict sense of the word, there's certainly a fractal quality to it.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"As an added bonus, it happens to look like a bunch of popsicles.\"})})]}),/*#__PURE__*/e(\"h2\",{children:\"Different flower, same essence\"}),/*#__PURE__*/e(\"p\",{children:\"Toloka has gone a long way since its launch in 2014, and now we're bigger and stronger than ever. Looks aren't everything \u2013 we know that \u2013 but changing your appearance can help better represent who you are on the inside. We wanted our clients and users to see us the way we see ourselves: forward-looking, innovative, accessible, friendly. That's what we want our new look to achieve, and, frankly, we love it so much we're certain it will hit the mark.\"}),/*#__PURE__*/e(\"img\",{alt:\"New identity\",className:\"framer-image\",height:\"309\",src:\"https://framerusercontent.com/images/lBfsWwrAInKhYrIsLd6l8V2dE.jpeg\",srcSet:\"https://framerusercontent.com/images/lBfsWwrAInKhYrIsLd6l8V2dE.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/lBfsWwrAInKhYrIsLd6l8V2dE.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/lBfsWwrAInKhYrIsLd6l8V2dE.jpeg 1840w\",style:{aspectRatio:\"1840 / 618\"},width:\"920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"New identity\"})}),/*#__PURE__*/e(\"p\",{children:\"The best part is that we didn't have to completely tear down our old identity in favor of a new, more refined appearance. It's still us! We even managed to keep the flower in our logo. We tried (at one point, really hard) to get rid of it, but it didn't work. It took us a long cycle of planning, approving, and testing to realize that it's part of who we are.\"}),/*#__PURE__*/e(\"p\",{children:\"Whether you are a Toloker or a requester, we hope you enjoy the renewed Toloka as much as we do. It is a big step for us, and we are incredibly grateful that you are taking it with us.\"})]});export const richText13=/*#__PURE__*/t(n.Fragment,{children:[/*#__PURE__*/t(\"p\",{children:[\"Neural Information Processing Systems is an \",/*#__PURE__*/e(a,{href:\"https://nips.cc/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"annual conference\"})}),\" that featured more than 2,000 papers when it was last held in December 2020. \",/*#__PURE__*/e(a,{href:\"https://www.linkedin.com/in/adrutsa/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Alexey Drutsa\"})}),\" and \",/*#__PURE__*/e(a,{href:\"https://www.linkedin.com/in/alexeyvolobuev/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Alexey Volobuev\"})}),\" from the Toloka team have chosen six papers from the conference as essential reading in crowd science. They discussed their selection and shared their perspective on trends in \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/knowledgebase/crowdsourcing-concepts/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"crowdsourcing\"})}),\" research. In this post we'll focus on some general crowdsourcing aspects \u2014 for the full details, see the \",/*#__PURE__*/e(a,{href:\"https://www.youtube.com/watch?v=u1sYgLOyBDw\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"video\"})}),\" of the seminar. \"]}),/*#__PURE__*/e(\"p\",{children:\"Here are the top NeurIPS publications most relevant to crowd science:\"}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(a,{href:\"https://papers.nips.cc/paper/2020/hash/386854131f58a556343e056f03626e00-Abstract.html\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Second Order PAC-Bayesian Bounds for the Weighted Majority Vote\"})})})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(a,{href:\"https://papers.nips.cc/paper/2020/hash/f86890095c957e9b949d11d15f0d0cd5-Abstract.html\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Adversarial Crowdsourcing Through Robust Rank-One Matrix Completion\"})})})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(a,{href:\"https://papers.nips.cc/paper/2020/hash/460191c72f67e90150a093b4585e7eb4-Abstract.html\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Improving Natural Language Processing Tasks with Human Gaze-Guided Neural Attention\"})})})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(a,{href:\"https://papers.nips.cc/paper/2020/hash/1f89885d556929e98d3ef9b86448f951-Abstract.html\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Learning to Summarize with Human Feedback\"})})})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(a,{href:\"https://papers.nips.cc/paper/2020/hash/b5d17ed2b502da15aa727af0d51508d6-Abstract.html\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"Disentangling Human Error from Ground Truth in Segmentation of Medical Images\"})})})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(a,{href:\"https://papers.nips.cc/paper/2020/hash/f4e3ce3e7b581ff32e40968298ba013d-Abstract.html\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"A Topological Filter for Learning with Label Noise\"})})})})]}),/*#__PURE__*/e(\"h2\",{children:\"Second Order PAC-Bayesian Bounds for the Weighted Majority Vote\"}),/*#__PURE__*/e(\"p\",{children:\"Andres Masegosa, Stephan Lorenzen, Christian Igel, Yevgeny Seldin\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Overview\"})}),/*#__PURE__*/e(\"p\",{children:\"It may come as a surprise that majority vote is a fundamental technique used in robust state-of-the-art machine learning systems. In fact, the winning strategy at most machine learning competitions employs weighted majority vote as a way to combine different prediction models. Majority vote performs so well because it removes the effect of errors in predictions by averaging out the noise in individual predictions. When individual classifiers are of different quality, it is important to assign different weights to them when aggregating their predictions. This paper focuses on the problem of weight selection and develops a practical technique that does not require individual classifiers to be independent.\"}),/*#__PURE__*/e(\"img\",{alt:\"Weighted Majority Vote\",className:\"framer-image\",height:\"380\",src:\"https://framerusercontent.com/images/GNv1NXzmrEEcp0l1bLaDYXYBvE.jpeg\",srcSet:\"https://framerusercontent.com/images/GNv1NXzmrEEcp0l1bLaDYXYBvE.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/GNv1NXzmrEEcp0l1bLaDYXYBvE.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/GNv1NXzmrEEcp0l1bLaDYXYBvE.jpeg 1840w\",style:{aspectRatio:\"1840 / 760\"},width:\"920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Weighted Majority Vote\"})}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Key takeaway\"})}),/*#__PURE__*/e(\"p\",{children:\"Majority vote (MV) is a popular algorithm that is often used in crowdsourcing for aggregating results. However, it has a weakness: all responses are weighted equally when determining the majority, which doesn't reflect the huge variation in quality of performers. Poor performers may skew results. If we could give more weight to responses from highly-skilled performers, we could be more confident in the aggregated result. This paper offers a strategy for successfully selecting the weights to use in order to achieve better results.\"}),/*#__PURE__*/e(\"h2\",{children:\"Adversarial Crowdsourcing Through Robust Rank-One Matrix Completion\"}),/*#__PURE__*/e(\"p\",{children:\"Qianqian Ma, Alex Olshevsky\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Overview\"})}),/*#__PURE__*/e(\"p\",{children:\"The famous Dawid-Skene model is often used in crowdsourcing, where the key aspect is performer skill. Many different algorithms have been developed for estimating performer skill levels and using these predictions when aggregating responses. However, the model's assumptions are not always true in practice (in other words, not all individual performers demonstrate the predicted skill in real tasks), which means that standard methods are often unreliable.\"}),/*#__PURE__*/e(\"img\",{alt:\"Dawid-Skene Model\",className:\"framer-image\",height:\"380\",src:\"https://framerusercontent.com/images/AGzzzk5qqWuUNpvqVSgISIm66P4.jpeg\",srcSet:\"https://framerusercontent.com/images/AGzzzk5qqWuUNpvqVSgISIm66P4.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/AGzzzk5qqWuUNpvqVSgISIm66P4.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/AGzzzk5qqWuUNpvqVSgISIm66P4.jpeg 1840w\",style:{aspectRatio:\"1840 / 760\"},width:\"920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Dawid-Skene Model\"})}),/*#__PURE__*/e(\"p\",{children:\"This paper suggests a new method involving matrix completion in order to discover performer quality level from incomplete data. The authors believe their method is robust when performers deviate from the model, which means it can effectively improve the quality of skill predictions, and ultimately, aggregated results.\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Key takeaway\"})}),/*#__PURE__*/e(\"p\",{children:\"The proposed algorithm could be a successful method for aggregating results in scenarios with dishonest or low-quality performers. This method could be applied to crowdsourcing projects and used for creating new crowd science solutions.\"}),/*#__PURE__*/e(\"h2\",{children:\"Improving Natural Language Processing Tasks with Human Gaze-Guided Neural Attention\"}),/*#__PURE__*/e(\"p\",{children:\"Ekta Sood, Simon Tannert, Philipp Mueller, Andreas Bulling\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Overview\"})}),/*#__PURE__*/e(\"p\",{children:\"Modern computer vision models often use supervision with human gaze data to understand what parts of an image are most important and then use this information to improve the quality. Could models related to text processing also benefit from gathering alternative types of data? Unfortunately, there isn't much gaze data available for texts. The existing corpora usually contain small amounts of data related to which parts of a text receive more attention from a human and which receive less. Crowdsourcing is key to collecting new types of data to accelerate natural language processing.\"}),/*#__PURE__*/e(\"p\",{children:\"The authors propose the following solutions:\"}),/*#__PURE__*/t(\"ul\",{style:{\"--framer-font-size\":\"18px\",\"--framer-text-alignment\":\"start\",\"--framer-text-color\":\"rgb(30, 33, 38)\",\"--framer-text-stroke-width\":\"0px\",\"--framer-text-transform\":\"none\"},children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"A hybrid text saliency model where they combined human gaze data with a cognitive model known as EZ Reader.\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Joint training of parameters for the entire network updated for a target task.\"})})]}),/*#__PURE__*/e(\"img\",{alt:\"Supervision Using Human Attention\",className:\"framer-image\",height:\"380\",src:\"https://framerusercontent.com/images/UCGSvL88knggWicEppLDAFA09Cs.jpeg\",srcSet:\"https://framerusercontent.com/images/UCGSvL88knggWicEppLDAFA09Cs.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/UCGSvL88knggWicEppLDAFA09Cs.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/UCGSvL88knggWicEppLDAFA09Cs.jpeg 1840w\",style:{aspectRatio:\"1840 / 760\"},width:\"920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Supervision Using Human Attention\"})}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Key takeaway\"})}),/*#__PURE__*/t(\"p\",{children:[\"Natural language processing is an important field in machine learning, yet it still experiences a lack of \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/data-labeling-platform/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"human-labeled data\"})}),\". Models could be significantly improved by providing them with large quantities of data and using new types of data for training. In addition, future research could involve human-in-the-loop technology for training parameters.\"]}),/*#__PURE__*/e(\"h2\",{children:\"Learning to Summarize with Human Feedback\"}),/*#__PURE__*/e(\"p\",{children:\"Nisan Stiennon, Long Ouyang, Jeffrey Wu, Daniel Ziegler, Ryan Lowe, Chelsea Voss, Alec Radford, Dario Amodei, Paul F. Christiano\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Overview\"})}),/*#__PURE__*/e(\"p\",{children:\"This paper looks at the challenges of summarization, which is a natural language processing (NLP) task that takes a long and complicated text and extracts a short summary. Large transformer networks like BERT and GPT-3 achieve good results, but optimization is limited because standard text quality metrics like ROUGE don\u2019t accurately reflect the quality of a summary and often lead to various artifacts.\"}),/*#__PURE__*/e(\"p\",{children:\"The solution is to collect human feedback: crowd performers choose which summaries are better, and this information is used to train a reward model and a policy that will generate high-quality summaries.\"}),/*#__PURE__*/e(\"img\",{alt:\"Human feedback techniques\",className:\"framer-image\",height:\"723\",src:\"https://framerusercontent.com/images/dVEmHZYX9bcJCKUFVmwbxoYoJLs.webp\",srcSet:\"https://framerusercontent.com/images/dVEmHZYX9bcJCKUFVmwbxoYoJLs.webp?scale-down-to=512 512w,https://framerusercontent.com/images/dVEmHZYX9bcJCKUFVmwbxoYoJLs.webp?scale-down-to=1024 1024w,https://framerusercontent.com/images/dVEmHZYX9bcJCKUFVmwbxoYoJLs.webp?scale-down-to=2048 2048w,https://framerusercontent.com/images/dVEmHZYX9bcJCKUFVmwbxoYoJLs.webp 2560w\",style:{aspectRatio:\"2560 / 1446\"},width:\"1280\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Human feedback techniques \"})}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Key takeaway\"})}),/*#__PURE__*/e(\"p\",{children:\"Human feedback techniques can help models improve performance on practical NLP tasks. Crowdsourcing plays a crucial role in human-in-the-loop pipelines.\"}),/*#__PURE__*/e(\"h2\",{children:\"Disentangling Human Error from the Ground Truth in Segmentation of Medical Images\"}),/*#__PURE__*/e(\"p\",{children:\"Le Zhang, Ryutaro Tanno, Mou-Cheng Xu, Chen Jin, Joseph Jacob, Olga Cicarrelli, Frederik Barkhof, Daniel Alexander\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Overview\"})}),/*#__PURE__*/e(\"p\",{children:\"Tasks that require selecting areas in complex images, like medical scans, often show a great deal of variation in answers. It can be difficult to establish the ground truth because even professional experts are inconsistent, and this problem creates obstacles for training AI. In this paper the authors propose a solution for using human-provided answers to train automatic segmentation algorithms, without getting derailed by flawed data.\"}),/*#__PURE__*/e(\"img\",{alt:\"Proposed solution\",className:\"framer-image\",height:\"723\",src:\"https://framerusercontent.com/images/r7EwUUNFAaqhNb0bnpZ136MU.webp\",srcSet:\"https://framerusercontent.com/images/r7EwUUNFAaqhNb0bnpZ136MU.webp?scale-down-to=512 512w,https://framerusercontent.com/images/r7EwUUNFAaqhNb0bnpZ136MU.webp?scale-down-to=1024 1024w,https://framerusercontent.com/images/r7EwUUNFAaqhNb0bnpZ136MU.webp?scale-down-to=2048 2048w,https://framerusercontent.com/images/r7EwUUNFAaqhNb0bnpZ136MU.webp 2560w\",style:{aspectRatio:\"2560 / 1446\"},width:\"1280\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Proposed solution \"})}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Key takeaway\"})}),/*#__PURE__*/e(\"p\",{children:\"Jointly training a model and learning how to aggregate data looks very promising \u2014 doing this with a creative probabilistic model could achieve state-of-the-art results in the medical segmentation field. Extending this method to other tasks may be a promising area of research, resulting in new cost-effective ways to train models while using noisy data.\"}),/*#__PURE__*/e(\"h2\",{children:\"A Topological Filter for Learning with Label Noise\"}),/*#__PURE__*/e(\"p\",{children:\"Pengxiang Wu, Songzhu Zheng, Mayank Goswami, Dimitris Metaxas, Chao Chen\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Overview\"})}),/*#__PURE__*/e(\"p\",{children:\"If you send your data to the crowd for labeling, how will you know whether the resulting labeled dataset contains any errors? What if you could detect errors and fix them or remove them from the dataset before training? This paper offers anew way to do this with a robust method for filtering noise and collecting clean data.\"}),/*#__PURE__*/e(\"img\",{alt:\"Clean and corrupted data\",className:\"framer-image\",height:\"380\",src:\"https://framerusercontent.com/images/DvUFFb8RlsrGTu5ia1t6uhUCQak.jpeg\",srcSet:\"https://framerusercontent.com/images/DvUFFb8RlsrGTu5ia1t6uhUCQak.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/DvUFFb8RlsrGTu5ia1t6uhUCQak.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/DvUFFb8RlsrGTu5ia1t6uhUCQak.jpeg 1840w\",style:{aspectRatio:\"1840 / 760\"},width:\"920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Clean and corrupted data\"})}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Key takeaway\"})}),/*#__PURE__*/e(\"p\",{children:\"Deep learning models are good at finding noisy labels because they interfere with learning. Previous research mostly focused on exploiting uncertain labels. This method leverages the group behavior of data in the latent representation, which has been neglected by previous classifier-confidence-dependent approaches.\"})]});export const richText14=/*#__PURE__*/t(n.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"Hundreds of thousands of people go missing every year worldwide, and there are many unseen heroes working hard to prevent tragic endings to their stories. At Toloka, we know how technology can benefit people in communication, business, and quality of life \u2014 and we also believe that technology can help bring people home. Toloka now offers a way to contribute to search efforts: we have teamed up with the Liza Alert volunteer search-and-rescue team to find missing people. As part of the process, Tolokers can check aerial images for signs of people to boost searches in progress. \"}),/*#__PURE__*/e(\"h2\",{children:\"Liza Alert's mission\"}),/*#__PURE__*/e(\"p\",{children:\"Liza Alert is a nonprofit organization that coordinates volunteers to search for missing people in Russia and neighboring countries of the former USSR, with an alert system similar to Amber Alert in the United States and Europe.\"}),/*#__PURE__*/e(\"p\",{children:\"Liza Alert searches for anyone who has been reported missing in cities or in the wilderness, with priority given to children and the elderly. In the 10 years since Liza Alert was founded, over 100,000 searches have been carried out and more than 85,000 missing people have been found. In locations that are difficult to access on foot, the organization uses drones to capture photos.\"}),/*#__PURE__*/e(\"h2\",{children:\"Optimizing the search effort\"}),/*#__PURE__*/e(\"p\",{children:\"The new collaborative project optimizes searches with image labeling. Tolokers will look at aerial images photographed by drones and mark the ones where they see people.\"}),/*#__PURE__*/e(\"img\",{alt:\"Task example\",className:\"framer-image\",height:\"946\",src:\"https://framerusercontent.com/images/Lpbr9NNCvEUBEckIvC7l0LI84Eo.webp\",srcSet:\"https://framerusercontent.com/images/Lpbr9NNCvEUBEckIvC7l0LI84Eo.webp?scale-down-to=512 512w,https://framerusercontent.com/images/Lpbr9NNCvEUBEckIvC7l0LI84Eo.webp?scale-down-to=1024 1024w,https://framerusercontent.com/images/Lpbr9NNCvEUBEckIvC7l0LI84Eo.webp?scale-down-to=2048 2048w,https://framerusercontent.com/images/Lpbr9NNCvEUBEckIvC7l0LI84Eo.webp 2560w\",style:{aspectRatio:\"2560 / 1893\"},width:\"1280\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Task example\"})}),/*#__PURE__*/e(\"p\",{children:\"Experts will then analyze the photos with people in them and decide whether to send a search-and-rescue team to the location. To ensure the quality of results and reduce the number of errors, each photo will be labeled by multiple users. Results from different users will be compared and aggregated.\"}),/*#__PURE__*/e(\"p\",{children:\"Toloka plans to review up to 250,000 images per month. In comparison, Liza Alert volunteers managed to check just under 700,000 drone images during the year 2020. By integrating this process in Toloka, the organization can now access a crowdforce with hundreds of people available to check images at the same time. This can dramatically speed up image processing and free up volunteers to focus on more specialized tasks.\"}),/*#__PURE__*/e(\"h2\",{children:\"Saving time and saving lives\"}),/*#__PURE__*/t(\"p\",{children:[\"For people lost in the wilderness, every minute counts. A combination of drones and \",/*#__PURE__*/e(a,{href:\"https://toloka.ai/knowledgebase/crowdsourcing-concepts/\",motionChild:!0,nodeId:\"eo4RAmtig\",openInNewTab:!1,scopeId:\"contentManagement\",smoothScroll:!1,children:/*#__PURE__*/e(o.a,{children:\"crowdsourcing\"})}),\" can streamline the search process and save precious time in getting volunteers to the right location. The images labeled in Toloka will eventually be collected as a dataset and used for training AI. For Tolokers, the new project is a unique opportunity to make an impact and help save lives.\"]})]});export const richText15=/*#__PURE__*/t(n.Fragment,{children:[/*#__PURE__*/e(\"h2\",{children:\"Who are Tolokers?\"}),/*#__PURE__*/e(\"p\",{children:\"The Toloka crowdsourcing platform helps process large amounts of data by distributing tasks to a large pool of users \u2014 we call them Tolokers. \"}),/*#__PURE__*/e(\"p\",{children:\"Tolokers can transcribe audio recordings, complete focus group surveys, moderate comments, classify images, and much more. The resulting data can be used for training machine learning algorithms or solving other business tasks.\"}),/*#__PURE__*/e(\"p\",{children:\"The platform works like a marketplace where two sides come together: requesters (individuals or companies) post tasks and set a price, while Tolokers choose tasks they like and do them in their spare time. Toloka has over 9 million registered users, with more than 40,000 Tolokers completing an average of 15 million tasks every day.\"}),/*#__PURE__*/e(\"p\",{children:\"Sometimes requesters want to know more about the users doing their tasks, like age, gender, or occupation. On the other hand, people who are considering becoming Tolokers may also be interested in learning about other people who choose to earn money on the platform \u2014 and whether it's worth it. To satisfy the curiosity of both sides, we ran a survey to learn more about the Toloka community.\"}),/*#__PURE__*/e(\"p\",{children:\"We asked Tolokers about their education, employment, and hobbies, and gathered responses from over 500,000 Tolokers in 191 countries who speak 186 languages. Here's what we discovered.\"}),/*#__PURE__*/e(\"h2\",{children:\"Age, gender and education\"}),/*#__PURE__*/e(\"p\",{children:\"As for the gender ratio, 58% of respondents were male and 42% were female. Age-wise, the majority of Tolokers are in their prime, with about half under age 25.\"}),/*#__PURE__*/e(\"img\",{alt:\"Age and gender\",className:\"framer-image\",height:\"411\",src:\"https://framerusercontent.com/images/bxTNqInFWEOKIC9VxrUivT5Dk7Q.jpeg\",srcSet:\"https://framerusercontent.com/images/bxTNqInFWEOKIC9VxrUivT5Dk7Q.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/bxTNqInFWEOKIC9VxrUivT5Dk7Q.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/bxTNqInFWEOKIC9VxrUivT5Dk7Q.jpeg 1840w\",style:{aspectRatio:\"1840 / 822\"},width:\"920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Age and gender\"})}),/*#__PURE__*/e(\"p\",{children:\"Youthfulness aside, most have already completed higher education: 40% have earned a degree, and 27% have vocational training. Both STEM and humanities majors are looking for extra earnings on the platform, contrary to many popular assumptions about the demand for tech careers.\"}),/*#__PURE__*/e(\"img\",{alt:\"Education\",className:\"framer-image\",height:\"483\",src:\"https://framerusercontent.com/images/9rc1DJkfmdWcb16OVZXDShGfPQ.jpeg\",srcSet:\"https://framerusercontent.com/images/9rc1DJkfmdWcb16OVZXDShGfPQ.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/9rc1DJkfmdWcb16OVZXDShGfPQ.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/9rc1DJkfmdWcb16OVZXDShGfPQ.jpeg 1840w\",style:{aspectRatio:\"1840 / 966\"},width:\"920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Education\"})}),/*#__PURE__*/e(\"h2\",{children:\"Employment and professions\"}),/*#__PURE__*/e(\"p\",{children:\"Most Tolokers have a source of income outside of Toloka. The largest group of users (35%) have a full-time job, 17% are focused on studies, and 12% are on maternity leave. For the vast majority, Toloka provides a way to earn some extra cash without set time commitments.\"}),/*#__PURE__*/e(\"img\",{alt:\"Employment\",className:\"framer-image\",height:\"483\",src:\"https://framerusercontent.com/images/9rc1DJkfmdWcb16OVZXDShGfPQ.jpeg\",srcSet:\"https://framerusercontent.com/images/9rc1DJkfmdWcb16OVZXDShGfPQ.jpeg?scale-down-to=512 512w,https://framerusercontent.com/images/9rc1DJkfmdWcb16OVZXDShGfPQ.jpeg?scale-down-to=1024 1024w,https://framerusercontent.com/images/9rc1DJkfmdWcb16OVZXDShGfPQ.jpeg 1840w\",style:{aspectRatio:\"1840 / 966\"},width:\"920\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Employment\"})}),/*#__PURE__*/e(\"p\",{children:\"Let's look at what kind of jobs these users have. Women most often work in retail (21%) and education (12%) and a significant number of men are engineers (17%). Other popular professions are construction and accounting, but there are users who identify themselves as pilots, mechanics, administrators, miners, movers, designers, warehouse workers, woodworkers, realtors, marketing specialists, chemists, seamstresses, sailors, and even jewelers (who probably handle tasks with the utmost delicacy).\"}),/*#__PURE__*/e(\"img\",{alt:\"Professions\",className:\"framer-image\",height:\"566\",src:\"https://framerusercontent.com/images/YfofSoKU6Sb0HImgsGkbXdbyBQU.webp\",srcSet:\"https://framerusercontent.com/images/YfofSoKU6Sb0HImgsGkbXdbyBQU.webp?scale-down-to=512 512w,https://framerusercontent.com/images/YfofSoKU6Sb0HImgsGkbXdbyBQU.webp?scale-down-to=1024 1024w,https://framerusercontent.com/images/YfofSoKU6Sb0HImgsGkbXdbyBQU.webp?scale-down-to=2048 2048w,https://framerusercontent.com/images/YfofSoKU6Sb0HImgsGkbXdbyBQU.webp 2560w\",style:{aspectRatio:\"2560 / 1133\"},width:\"1280\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Professions\"})}),/*#__PURE__*/e(\"h2\",{children:\"Hobbies\"}),/*#__PURE__*/e(\"p\",{children:\"When asked about their spare time, most Toloka users responded that they spend it online. They also spend their time doing chores, hitting the town, and hanging out with friends. Most prefer books over TV and real sports over e-sports.\"}),/*#__PURE__*/e(\"img\",{alt:\"Hobbies\",className:\"framer-image\",height:\"566\",src:\"https://framerusercontent.com/images/tKHncbNrQEXNFPc4Y3xUTAnp0.webp\",srcSet:\"https://framerusercontent.com/images/tKHncbNrQEXNFPc4Y3xUTAnp0.webp?scale-down-to=512 512w,https://framerusercontent.com/images/tKHncbNrQEXNFPc4Y3xUTAnp0.webp?scale-down-to=1024 1024w,https://framerusercontent.com/images/tKHncbNrQEXNFPc4Y3xUTAnp0.webp?scale-down-to=2048 2048w,https://framerusercontent.com/images/tKHncbNrQEXNFPc4Y3xUTAnp0.webp 2560w\",style:{aspectRatio:\"2560 / 1133\"},width:\"1280\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"em\",{children:\"Hobbies\"})}),/*#__PURE__*/e(\"p\",{children:\"A few users reported less common pastimes: home remodeling, repairing shoes, studying electronic engineering, voice-overs for videos and streams, bee keeping, crypto trading, church, making electric guitars, growing hot peppers, long commutes, or seeking enlightenment. Some users didn't identify any standard hobbies. A quarter of these people focus on their children, while others spend most of their free time job hunting, working (they just can't take a break!), snoring away in bed, or devoting themselves to Toloka \u2014 a prudent choice. \"}),/*#__PURE__*/e(\"h2\",{children:\"Portrait of a Toloker\"}),/*#__PURE__*/e(\"p\",{children:\"So what did we learn at the end of the day?\"}),/*#__PURE__*/e(\"p\",{children:\"A portrait of the average Toloker shows a young, educated individual with a day job in retail, engineering, or tech who spends their evenings online and uses weekends to run errands and relax with friends. These people view Toloka as a great way to kill time before or after work, or maybe on their commute. Other groups, like students and stay-at-home parents, have the flexibility to fit Toloka into their busy days as the opportunity arises.\"}),/*#__PURE__*/e(\"p\",{children:\"Tolokers come from a myriad of cultural and language backgrounds around the world. We have identified some global trends across Toloka users, but it is also fascinating to consider how each individual has a unique combination of knowledge, experience and interests to contribute to the global crowd.\"})]});\nexport const __FramerMetadata__ = {\"exports\":{\"richText1\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText2\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText11\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText8\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText3\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText10\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText12\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText13\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText5\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText14\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText15\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText7\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText4\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText9\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText6\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"__FramerMetadata__\":{\"type\":\"variable\"}}}"],
  "mappings": "6JAAsJ,IAAMA,EAAsBC,EAAIC,EAAS,CAAC,SAAS,CAAcC,EAAE,IAAI,CAAC,SAAS,+SAA+S,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,kBAAkB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yTAAyT,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,mBAAmB,UAAU,eAAe,OAAO,MAAM,IAAI,uEAAuE,OAAO,qWAAqW,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qQAAqQ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gGAAgG,CAAC,EAAeF,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,mBAAmB,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,gDAAgD,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,0BAA0B,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oQAAoQ,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,wCAAwC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,oIAAiJE,EAAE,KAAK,CAAC,SAAS,KAAK,CAAC,EAAE,QAAqBA,EAAE,KAAK,CAAC,SAAS,SAAS,CAAC,EAAE,+hBAA+hB,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,+UAA8VE,EAAE,KAAK,CAAC,SAAS,OAAO,CAAC,EAAE,QAAqBA,EAAE,KAAK,CAAC,SAAS,KAAK,CAAC,EAAE,4IAAyJA,EAAE,KAAK,CAAC,SAAS,OAAO,CAAC,EAAE,QAAqBA,EAAE,KAAK,CAAC,SAAS,KAAK,CAAC,EAAE,gEAAgE,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mQAAmQ,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,6BAA6B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,wcAA4c,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oXAAoX,CAAC,EAAeF,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,0BAA0B,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,gDAAgD,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,oEAAoE,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,gEAA6EE,EAAE,KAAK,CAAC,SAAS,sIAAsI,CAAC,EAAE,GAAG,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4YAA4Y,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,kBAAkB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qJAAqJ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mEAAmE,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,mBAAmB,UAAU,eAAe,OAAO,MAAM,IAAI,qEAAqE,OAAO,6VAA6V,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4IAA4I,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,+DAA+D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8DAA8D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oTAAoT,CAAC,CAAC,CAAC,CAAC,EAAeC,EAAuBH,EAAIC,EAAS,CAAC,SAAS,CAAcC,EAAE,KAAK,CAAC,SAAS,kDAAkD,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,kSAA+SE,EAAEE,EAAE,CAAC,KAAK,qGAAqG,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,iDAAiD,CAAC,CAAC,CAAC,EAAE,8DAA8D,CAAC,CAAC,EAAeL,EAAE,IAAI,CAAC,SAAS,CAAC,gLAA6LE,EAAEE,EAAE,CAAC,KAAK,4HAA4H,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,6BAA6B,CAAC,CAAC,CAAC,EAAE,kPAAkP,CAAC,CAAC,EAAeH,EAAE,IAAI,CAAC,SAAS,iFAAiF,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uDAAuD,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kDAAkD,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,kBAAkB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mZAAmZ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sQAAsQ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8NAA8N,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,sBAAsB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qJAAqJ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,wWAAwW,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,0BAA0B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8VAA8V,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,sBAAsB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sRAAiR,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8XAA8X,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sGAAsG,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,UAAU,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2YAA2Y,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,sBAAsB,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,2JAAwKE,EAAEE,EAAE,CAAC,KAAK,0DAA0D,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,uCAAuC,CAAC,CAAC,EAAeH,EAAE,IAAI,CAAC,SAAS,kcAAkc,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,4BAAyCE,EAAEE,EAAE,CAAC,KAAK,4CAA4C,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,omBAAqmB,CAAC,CAAC,EAAeH,EAAE,IAAI,CAAC,SAAS,kRAAkR,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,uCAAuC,UAAU,eAAe,OAAO,MAAM,IAAI,sEAAsE,OAAO,oQAAoQ,MAAM,CAAC,YAAY,YAAY,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,UAAU,gBAAgB,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeI,EAAuBN,EAAIC,EAAS,CAAC,SAAS,CAAcC,EAAE,IAAI,CAAC,SAAS,4QAAuQ,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,WAAW,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mOAAmO,CAAC,EAAeF,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,0FAAqF,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,6CAA6C,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,yGAAyG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,gLAAwLE,EAAEE,EAAE,CAAC,KAAK,4CAA4C,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,cAAc,CAAC,CAAC,CAAC,EAAE,sLAAsL,CAAC,CAAC,EAAeH,EAAE,IAAI,CAAC,SAAS,uVAAkV,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,wDAAwD,CAAC,EAAeF,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,2CAA2C,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,wEAAwE,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,qFAAqF,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,UAAU,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4FAA4F,CAAC,EAAeF,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,qGAAgG,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,wGAAmG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,WAAW,UAAU,eAAe,OAAO,MAAM,IAAI,sEAAsE,OAAO,iWAAiW,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,iNAAyNE,EAAEE,EAAE,CAAC,KAAK,4CAA4C,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,iBAAiB,CAAC,CAAC,EAAeH,EAAE,MAAM,CAAC,IAAI,mCAAmC,UAAU,eAAe,OAAO,MAAM,IAAI,uEAAuE,OAAO,uQAAuQ,MAAM,CAAC,YAAY,YAAY,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oDAAoD,CAAC,EAAeF,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,uBAAuB,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,iBAAiB,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,iBAAiB,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,6BAA6B,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8QAA8Q,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,SAAS,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,imBAAulB,CAAC,CAAC,CAAC,CAAC,EAAeK,EAAuBP,EAAIC,EAAS,CAAC,SAAS,CAAcD,EAAE,IAAI,CAAC,SAAS,CAAC,gOAAwOE,EAAEE,EAAE,CAAC,KAAK,8DAA8D,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,yGAAyG,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,SAAS,yBAAyB,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAEE,EAAE,CAAC,KAAK,6DAA6D,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,iMAAyMH,EAAEE,EAAE,CAAC,KAAK,oCAAoC,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,OAAO,CAAC,CAAC,CAAC,EAAE,sDAAmEH,EAAEE,EAAE,CAAC,KAAK,8BAA8B,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,OAAO,CAAC,CAAC,CAAC,EAAE,8EAA8E,CAAC,CAAC,EAAeH,EAAE,IAAI,CAAC,SAAS,wfAAwf,CAAC,EAAeF,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,iBAAiB,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,uBAAuB,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,uBAAuB,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,qDAAqD,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,yBAAyB,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,+SAAuTE,EAAEE,EAAE,CAAC,KAAK,yBAAyB,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,MAAM,CAAC,CAAC,CAAC,EAAE,0DAAuEH,EAAEE,EAAE,CAAC,KAAK,wBAAwB,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,QAAQ,CAAC,CAAC,CAAC,EAAE,gEAA6EH,EAAEE,EAAE,CAAC,KAAK,0BAA0B,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,OAAO,CAAC,CAAC,CAAC,EAAE,GAAG,CAAC,CAAC,EAAeH,EAAE,IAAI,CAAC,SAAS,mXAAyW,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,wBAAwB,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,0EAAuFE,EAAEE,EAAE,CAAC,KAAK,6BAA6B,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,iBAAiB,CAAC,CAAC,CAAC,EAAE,oKAA0J,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeG,EAAuBR,EAAIC,EAAS,CAAC,SAAS,CAAcC,EAAE,IAAI,CAAC,SAAS,sXAAiX,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,gOAA6OE,EAAEE,EAAE,CAAC,KAAK,4CAA4C,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,cAAc,CAAC,CAAC,CAAC,EAAE,sEAA8EH,EAAEE,EAAE,CAAC,KAAK,0DAA0D,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,+BAA+B,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,SAAS,sBAAsB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0cAA0c,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,oCAAoC,UAAU,eAAe,OAAO,OAAO,IAAI,wEAAwE,OAAO,4QAA4Q,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,oCAAoC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sEAAsE,CAAC,EAAeF,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,mOAAmO,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,sJAAiJ,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,uSAA+SE,EAAEE,EAAE,CAAC,KAAK,wCAAwC,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,WAAW,CAAC,CAAC,CAAC,EAAE,uCAAuC,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,SAAS,WAAW,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uKAAuK,CAAC,EAAeF,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,cAAc,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,aAAa,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,iDAAiD,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,MAAM,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,iBAAiB,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,yBAAyB,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,+GAA+G,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,0CAA0C,UAAU,eAAe,OAAO,MAAM,IAAI,wEAAwE,OAAO,0QAA0Q,MAAM,CAAC,YAAY,YAAY,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,yCAAyC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,iIAAuH,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,+WAA+W,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gOAA2N,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,4DAA4D,UAAU,eAAe,OAAO,MAAM,IAAI,uEAAuE,OAAO,uQAAuQ,MAAM,CAAC,YAAY,YAAY,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,kHAAwG,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oKAAoK,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,UAAU,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0pBAAqpB,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,iDAAiD,UAAU,eAAe,OAAO,MAAM,IAAI,uEAAuE,OAAO,uQAAuQ,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,oIAAoI,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0DAAqD,CAAC,EAAeF,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,8NAA8N,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,yHAAyH,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,sKAAsK,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qMAAqM,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,SAAS,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kbAA6a,CAAC,CAAC,CAAC,CAAC,EAAeO,EAAuBT,EAAIC,EAAS,CAAC,SAAS,CAAcD,EAAE,IAAI,CAAC,SAAS,CAAC,yJAA4JE,EAAEE,EAAE,CAAC,KAAK,4CAA4C,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,kGAA+GH,EAAEE,EAAE,CAAC,KAAK,yCAAyC,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,MAAM,CAAC,CAAC,CAAC,EAAE,QAAqBH,EAAEE,EAAE,CAAC,KAAK,2CAA2C,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,MAAM,CAAC,CAAC,CAAC,EAAE,IAAI,CAAC,CAAC,EAAeH,EAAE,IAAI,CAAC,SAAS,8LAA+K,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,+BAA+B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2JAAiJ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uKAAwJ,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,kBAAkB,UAAU,eAAe,OAAO,MAAM,IAAI,wEAAwE,OAAO,yWAAyW,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,uCAAuC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mNAAmN,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sFAAsF,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,mEAAmE,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oQAA0P,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,iSAAiS,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,4CAA4C,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qHAAgH,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8LAA8L,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,4CAA4C,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yQAAoQ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6KAAwK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ifAAue,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,2DAA2D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,wIAAmI,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kQAA6P,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,4EAA4E,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gOAAgO,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oQAA+P,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,0DAA0D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gMAA2L,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,wRAAoQ,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,kDAAkD,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2RAA4Q,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gRAAiQ,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,kEAAkE,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,wVAAyU,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,wRAAsRE,EAAEE,EAAE,CAAC,KAAK,0DAA0D,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,iOAA4N,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeK,EAAuBV,EAAIC,EAAS,CAAC,SAAS,CAAcD,EAAE,IAAI,CAAC,SAAS,CAAC,8BAA2CE,EAAEE,EAAE,CAAC,KAAK,mCAAmC,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,qDAAqD,CAAC,CAAC,CAAC,EAAE,kPAA8O,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,SAAS,oBAAoB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0OAA0O,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,6JAA0KE,EAAEE,EAAE,CAAC,KAAK,4CAA4C,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,sGAAsG,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,SAAS,UAAU,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,WAAW,UAAU,eAAe,OAAO,MAAM,IAAI,uEAAuE,OAAO,qWAAqW,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,UAAU,CAAC,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,iFAA8FE,EAAEE,EAAE,CAAC,KAAK,0DAA0D,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,GAAG,CAAC,CAAC,EAAeL,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAEE,EAAE,CAAC,KAAK,2CAA2C,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,iBAAiB,CAAC,CAAC,CAAC,EAAE,wMAAyM,CAAC,CAAC,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAEE,EAAE,CAAC,KAAK,kDAAkD,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,aAAa,CAAC,CAAC,CAAC,EAAE,uOAAuO,CAAC,CAAC,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAEE,EAAE,CAAC,KAAK,0CAA0C,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,iBAAiB,CAAC,CAAC,CAAC,EAAE,2PAA4P,CAAC,CAAC,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAEE,EAAE,CAAC,KAAK,yCAAyC,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,cAAc,CAAC,CAAC,CAAC,EAAE,mTAAmT,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,SAAS,MAAM,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,2JAAwKE,EAAEE,EAAE,CAAC,KAAK,uCAAuC,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,gBAAgB,CAAC,CAAC,CAAC,EAAE,QAAqBH,EAAEE,EAAE,CAAC,KAAK,uDAAuD,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,gBAAgB,CAAC,CAAC,CAAC,EAAE,oCAAiDH,EAAEE,EAAE,CAAC,KAAK,oDAAoD,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,aAAa,CAAC,CAAC,CAAC,EAAE,uJAAuJ,CAAC,CAAC,EAAeL,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBF,EAAE,IAAI,CAAC,SAAS,CAAC,iDAA8DE,EAAEE,EAAE,CAAC,KAAK,wCAAwC,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,WAAW,CAAC,CAAC,CAAC,EAAE,oCAAoC,CAAC,CAAC,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBF,EAAE,IAAI,CAAC,SAAS,CAAC,2FAAwGE,EAAEE,EAAE,CAAC,KAAK,uCAAuC,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,YAAY,CAAC,CAAC,CAAC,EAAE,oCAAoC,CAAC,CAAC,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBF,EAAE,IAAI,CAAC,SAAS,CAAC,OAAoBE,EAAEE,EAAE,CAAC,KAAK,sCAAsC,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,WAAW,CAAC,CAAC,CAAC,EAAE,4FAA4F,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeH,EAAE,IAAI,CAAC,SAAS,oHAAoH,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAEE,EAAE,CAAC,KAAK,6CAA6C,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,YAAY,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeM,EAAuBX,EAAIC,EAAS,CAAC,SAAS,CAAcD,EAAE,IAAI,CAAC,SAAS,CAAC,ubAAocE,EAAEE,EAAE,CAAC,KAAK,0DAA0D,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,2DAA2D,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,SAAS,gCAAgC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oFAAoF,CAAC,EAAeF,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,mDAAmD,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,wEAAwE,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,iDAAiD,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,uBAAuB,UAAU,eAAe,OAAO,MAAM,IAAI,uEAAuE,OAAO,uQAAuQ,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,sBAAsB,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qRAAqR,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gVAAgV,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,2BAA2B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4KAA4K,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,4BAA4B,UAAU,eAAe,OAAO,MAAM,IAAI,uEAAuE,OAAO,uQAAuQ,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,2BAA2B,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4KAA4K,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,eAAe,UAAU,eAAe,OAAO,MAAM,IAAI,wEAAwE,OAAO,0QAA0Q,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,cAAc,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uMAAuM,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,+OAA4PE,EAAE,KAAK,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,UAAU,gBAAgB,CAAC,CAAC,CAAC,CAAC,EAAeF,EAAE,KAAK,CAAC,SAAS,CAAcE,EAAE,KAAK,CAAC,CAAC,EAAE,SAAS,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2kBAAskB,CAAC,CAAC,CAAC,CAAC,EAAeU,EAAuBZ,EAAIC,EAAS,CAAC,SAAS,CAAcD,EAAE,IAAI,CAAC,SAAS,CAAC,uDAAoEE,EAAEE,EAAE,CAAC,KAAK,qCAAqC,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,aAAa,CAAC,CAAC,CAAC,EAAE,qUAAqU,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,SAAS,oBAAoB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,scAA4b,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,qQAAkRE,EAAEE,EAAE,CAAC,KAAK,0DAA0D,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,6GAA6G,CAAC,CAAC,EAAeH,EAAE,MAAM,CAAC,IAAI,SAAS,UAAU,eAAe,OAAO,MAAM,IAAI,uEAAuE,OAAO,uQAAuQ,MAAM,CAAC,YAAY,YAAY,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,QAAQ,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,qCAAqC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kUAAkU,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0dAA0d,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,aAAa,UAAU,eAAe,OAAO,MAAM,IAAI,uEAAuE,OAAO,qWAAqW,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,YAAY,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qNAAgN,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,mBAAmB,UAAU,eAAe,OAAO,MAAM,IAAI,uEAAuE,OAAO,uQAAuQ,MAAM,CAAC,YAAY,YAAY,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,kBAAkB,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oGAAoG,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,qBAAqB,UAAU,eAAe,OAAO,MAAM,IAAI,wEAAwE,OAAO,0QAA0Q,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,oBAAoB,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,UAAU,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,+UAA+U,CAAC,EAAeF,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,iMAAiM,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,kXAAkX,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,6OAA6O,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,WAAW,UAAU,eAAe,OAAO,MAAM,IAAI,uEAAuE,OAAO,uQAAuQ,MAAM,CAAC,YAAY,YAAY,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,UAAU,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,SAAS,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sRAAsR,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,UAAU,UAAU,eAAe,OAAO,MAAM,IAAI,wEAAwE,OAAO,0QAA0Q,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,SAAS,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeW,EAAuBb,EAAIC,EAAS,CAAC,SAAS,CAAcC,EAAE,IAAI,CAAC,SAAS,0OAA0O,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,0pBAAyqBE,EAAEE,EAAE,CAAC,KAAK,0DAA0D,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,wFAAwF,CAAC,CAAC,EAAeH,EAAE,MAAM,CAAC,IAAI,mBAAmB,UAAU,eAAe,OAAO,MAAM,IAAI,uEAAuE,OAAO,uQAAuQ,MAAM,CAAC,YAAY,YAAY,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,kBAAkB,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,+BAA+B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oPAA+O,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uNAAuN,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,iOAAiO,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,mEAAmE,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4fAA4f,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gZAAgZ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uYAA6X,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,4CAA4C,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8QAA8Q,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,4CAA4C,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,+PAA4QE,EAAEE,EAAE,CAAC,KAAK,4CAA4C,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,yuBAAyuB,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,SAAS,4EAA4E,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6iCAA6iC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,0DAA0D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,k8BAA67B,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,kDAAkD,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2tBAA2tB,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,kEAAkE,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,y0BAAy0B,CAAC,CAAC,CAAC,CAAC,EAAeY,EAAwBd,EAAIC,EAAS,CAAC,SAAS,CAAcD,EAAE,IAAI,CAAC,SAAS,CAAC,8SAA2TE,EAAEE,EAAE,CAAC,KAAK,0DAA0D,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,yDAAyD,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,SAAS,0CAA0C,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,+XAA0X,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6LAA6L,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,eAAe,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8FAA8F,CAAC,EAAeF,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,+EAA+E,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,+DAA+D,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,mBAAmB,UAAU,eAAe,OAAO,OAAO,IAAI,wEAAwE,OAAO,yWAAyW,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,kBAAkB,CAAC,CAAC,CAAC,EAAeF,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,8EAA8E,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,sPAAsP,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,mBAAmB,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,sCAAsC,UAAU,eAAe,OAAO,OAAO,IAAI,qEAAqE,OAAO,6VAA6V,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,qCAAqC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,SAAS,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oQAAoQ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4PAA4P,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8DAA8D,CAAC,EAAeF,EAAE,KAAK,CAAC,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,QAAQ,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,kBAAkB,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sEAAsE,CAAC,EAAeF,EAAE,KAAK,CAAC,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,iBAAiB,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,cAAc,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAea,EAAwBf,EAAIC,EAAS,CAAC,SAAS,CAAcC,EAAE,IAAI,CAAC,SAAS,uZAAkZ,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,uCAAuC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6VAAwV,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,wCAAwC,UAAU,eAAe,OAAO,OAAO,IAAI,sEAAsE,OAAO,qQAAqQ,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,SAAS,CAAC,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,wBAAwB,CAAC,EAAE,mTAAmT,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,wBAAwB,CAAC,EAAE,6KAA6K,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,sDAAsD,CAAC,EAAE,uOAAuO,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,iEAAiE,CAAC,EAAE,kUAAkU,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,mDAAmD,CAAC,EAAE,iNAAiN,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,kDAAkD,CAAC,EAAE,8YAAyY,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,wCAAwC,CAAC,EAAE,2FAAwGA,EAAEE,EAAE,CAAC,KAAK,sCAAsC,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,kBAAkB,CAAC,CAAC,CAAC,EAAE,2HAA2H,CAAC,CAAC,EAAeL,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,0DAA0D,CAAC,EAAE,6LAAwL,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,iFAAiF,CAAC,EAAE,mOAAoO,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,0BAA0B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,iJAAiJ,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,2BAA2B,UAAU,eAAe,OAAO,OAAO,IAAI,wEAAwE,OAAO,2QAA2Q,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,UAAU,CAAC,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,wBAAwB,CAAC,EAAE,2dAAsd,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,0BAA0B,CAAC,EAAE,8YAA8Y,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,iEAAiE,CAAC,EAAE,meAAme,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,uCAAuC,CAAC,EAAE,gCAA6CA,EAAEE,EAAE,CAAC,KAAK,4CAA4C,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,mKAAmK,CAAC,CAAC,EAAeL,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,iFAAiF,CAAC,EAAE,wSAAmS,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,gCAAgC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yLAAyL,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,iCAAiC,UAAU,eAAe,OAAO,OAAO,IAAI,sEAAsE,OAAO,qQAAqQ,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,OAAO,CAAC,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,8BAA8B,CAAC,EAAE,mRAAmR,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,0BAA0B,CAAC,EAAE,2hBAAwhB,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,+BAA+B,CAAC,EAAE,gNAAgN,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,iTAAiT,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6mBAA6mB,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,2CAA2C,CAAC,EAAE,kRAAkR,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,uCAAuC,CAAC,EAAE,6NAA6N,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,iFAAiF,CAAC,EAAE,gVAA2U,CAAC,CAAC,CAAC,CAAC,CAAC,EAAec,EAAwBhB,EAAIC,EAAS,CAAC,SAAS,CAAcC,EAAE,IAAI,CAAC,SAAS,ycAAyc,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,2BAA2B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gMAAgM,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,sEAAmFE,EAAEE,EAAE,CAAC,KAAK,0DAA0D,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,oDAAiEH,EAAEE,EAAE,CAAC,KAAK,4CAA4C,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,sQAAsQ,CAAC,CAAC,EAAeH,EAAE,MAAM,CAAC,IAAI,mBAAmB,UAAU,eAAe,OAAO,OAAO,IAAI,wEAAwE,OAAO,yWAAyW,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,kBAAkB,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uYAAkY,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4FAA4F,CAAC,EAAeF,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,8FAA8F,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,wHAAwH,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6SAA6S,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,eAAe,UAAU,eAAe,OAAO,OAAO,IAAI,sEAAsE,OAAO,iWAAiW,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,cAAc,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,oCAAoC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6RAA6R,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,WAAW,UAAU,eAAe,OAAO,OAAO,IAAI,wEAAwE,OAAO,yWAAyW,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,UAAU,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8BAA8B,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,WAAW,UAAU,eAAe,OAAO,OAAO,IAAI,wEAAwE,OAAO,yWAAyW,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,UAAU,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0MAA0M,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,GAAG,UAAU,eAAe,OAAO,MAAM,IAAI,uEAAuE,OAAO,wKAAwK,MAAM,CAAC,YAAY,WAAW,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,wKAAwK,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,WAAW,UAAU,eAAe,OAAO,MAAM,IAAI,wEAAwE,OAAO,0QAA0Q,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,UAAU,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,6BAA6B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,wbAAmb,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,gBAAgB,UAAU,eAAe,OAAO,MAAM,IAAI,uEAAuE,OAAO,qWAAqW,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gJAAgJ,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,gBAAgB,UAAU,eAAe,OAAO,OAAO,IAAI,wEAAwE,OAAO,yWAAyW,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8TAA8T,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,YAAY,UAAU,eAAe,OAAO,OAAO,IAAI,sEAAsE,OAAO,iWAAiW,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,WAAW,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kUAAkU,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,+PAA+P,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,WAAW,UAAU,eAAe,OAAO,OAAO,IAAI,wEAAwE,OAAO,yWAAyW,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,UAAU,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2FAA2F,CAAC,EAAeF,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,oEAAoE,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,gBAAgB,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,4GAA4G,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,kEAAkE,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,gCAAgC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,idAAuc,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,eAAe,UAAU,eAAe,OAAO,MAAM,IAAI,sEAAsE,OAAO,oQAAoQ,MAAM,CAAC,YAAY,YAAY,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,cAAc,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0WAA0W,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0LAA0L,CAAC,CAAC,CAAC,CAAC,EAAee,EAAwBjB,EAAIC,EAAS,CAAC,SAAS,CAAcD,EAAE,IAAI,CAAC,SAAS,CAAC,+CAA4DE,EAAEE,EAAE,CAAC,KAAK,mBAAmB,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,mBAAmB,CAAC,CAAC,CAAC,EAAE,iFAA8FH,EAAEE,EAAE,CAAC,KAAK,uCAAuC,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,QAAqBH,EAAEE,EAAE,CAAC,KAAK,8CAA8C,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,iBAAiB,CAAC,CAAC,CAAC,EAAE,oLAAiMH,EAAEE,EAAE,CAAC,KAAK,0DAA0D,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,kHAA0HH,EAAEE,EAAE,CAAC,KAAK,8CAA8C,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,OAAO,CAAC,CAAC,CAAC,EAAE,mBAAmB,CAAC,CAAC,EAAeH,EAAE,IAAI,CAAC,SAAS,uEAAuE,CAAC,EAAeF,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEE,EAAE,CAAC,KAAK,wFAAwF,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,iEAAiE,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEE,EAAE,CAAC,KAAK,wFAAwF,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,qEAAqE,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEE,EAAE,CAAC,KAAK,wFAAwF,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,qFAAqF,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEE,EAAE,CAAC,KAAK,wFAAwF,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,2CAA2C,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEE,EAAE,CAAC,KAAK,wFAAwF,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,+EAA+E,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAsBA,EAAEE,EAAE,CAAC,KAAK,wFAAwF,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,oDAAoD,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,SAAS,iEAAiE,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mEAAmE,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,UAAU,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0sBAA0sB,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,yBAAyB,UAAU,eAAe,OAAO,MAAM,IAAI,uEAAuE,OAAO,uQAAuQ,MAAM,CAAC,YAAY,YAAY,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,wBAAwB,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,cAAc,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yhBAAyhB,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,qEAAqE,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6BAA6B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,UAAU,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2cAA2c,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,oBAAoB,UAAU,eAAe,OAAO,MAAM,IAAI,wEAAwE,OAAO,0QAA0Q,MAAM,CAAC,YAAY,YAAY,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,mBAAmB,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,iUAAiU,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,cAAc,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8OAA8O,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,qFAAqF,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4DAA4D,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,UAAU,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8kBAA8kB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8CAA8C,CAAC,EAAeF,EAAE,KAAK,CAAC,MAAM,CAAC,qBAAqB,OAAO,0BAA0B,QAAQ,sBAAsB,kBAAkB,6BAA6B,MAAM,0BAA0B,MAAM,EAAE,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,6GAA6G,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,gFAAgF,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,oCAAoC,UAAU,eAAe,OAAO,MAAM,IAAI,wEAAwE,OAAO,0QAA0Q,MAAM,CAAC,YAAY,YAAY,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,mCAAmC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,cAAc,CAAC,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,6GAA0HE,EAAEE,EAAE,CAAC,KAAK,4CAA4C,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,oBAAoB,CAAC,CAAC,CAAC,EAAE,qOAAqO,CAAC,CAAC,EAAeH,EAAE,KAAK,CAAC,SAAS,2CAA2C,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kIAAkI,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,UAAU,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2ZAAsZ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6MAA6M,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,4BAA4B,UAAU,eAAe,OAAO,MAAM,IAAI,wEAAwE,OAAO,yWAAyW,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,4BAA4B,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,cAAc,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0JAA0J,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,mFAAmF,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oHAAoH,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,UAAU,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ybAAyb,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,oBAAoB,UAAU,eAAe,OAAO,MAAM,IAAI,qEAAqE,OAAO,6VAA6V,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,oBAAoB,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,cAAc,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yWAAoW,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,oDAAoD,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0EAA0E,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,UAAU,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uUAAuU,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,2BAA2B,UAAU,eAAe,OAAO,MAAM,IAAI,wEAAwE,OAAO,0QAA0Q,MAAM,CAAC,YAAY,YAAY,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,0BAA0B,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,cAAc,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8TAA8T,CAAC,CAAC,CAAC,CAAC,EAAegB,EAAwBlB,EAAIC,EAAS,CAAC,SAAS,CAAcC,EAAE,IAAI,CAAC,SAAS,6kBAAwkB,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,sBAAsB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sOAAsO,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,iYAAiY,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,8BAA8B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2KAA2K,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,eAAe,UAAU,eAAe,OAAO,MAAM,IAAI,wEAAwE,OAAO,yWAAyW,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,cAAc,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6SAA6S,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uaAAua,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,8BAA8B,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,uFAAoGE,EAAEE,EAAE,CAAC,KAAK,0DAA0D,YAAY,GAAG,OAAO,YAAY,aAAa,GAAG,QAAQ,oBAAoB,aAAa,GAAG,SAAsBF,EAAEG,EAAE,EAAE,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,sSAAsS,CAAC,CAAC,CAAC,CAAC,CAAC,EAAec,EAAwBnB,EAAIC,EAAS,CAAC,SAAS,CAAcC,EAAE,KAAK,CAAC,SAAS,mBAAmB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qJAAgJ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qOAAqO,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,+UAA+U,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,+YAA0Y,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0LAA0L,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,2BAA2B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,iKAAiK,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,iBAAiB,UAAU,eAAe,OAAO,MAAM,IAAI,wEAAwE,OAAO,0QAA0Q,MAAM,CAAC,YAAY,YAAY,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,gBAAgB,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uRAAuR,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,YAAY,UAAU,eAAe,OAAO,MAAM,IAAI,uEAAuE,OAAO,uQAAuQ,MAAM,CAAC,YAAY,YAAY,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,WAAW,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,4BAA4B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gRAAgR,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,aAAa,UAAU,eAAe,OAAO,MAAM,IAAI,uEAAuE,OAAO,uQAAuQ,MAAM,CAAC,YAAY,YAAY,EAAE,MAAM,KAAK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,YAAY,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ofAAof,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,cAAc,UAAU,eAAe,OAAO,MAAM,IAAI,wEAAwE,OAAO,yWAAyW,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,aAAa,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,SAAS,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6OAA6O,CAAC,EAAeA,EAAE,MAAM,CAAC,IAAI,UAAU,UAAU,eAAe,OAAO,MAAM,IAAI,sEAAsE,OAAO,iWAAiW,MAAM,CAAC,YAAY,aAAa,EAAE,MAAM,MAAM,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,KAAK,CAAC,SAAS,SAAS,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oiBAA+hB,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,uBAAuB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6CAA6C,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8bAA8b,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6SAA6S,CAAC,CAAC,CAAC,CAAC,EACpm1IkB,EAAqB,CAAC,QAAU,CAAC,UAAY,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,UAAY,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,WAAa,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,UAAY,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,UAAY,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,WAAa,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,WAAa,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,WAAa,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,UAAY,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,SAAW,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,WAAa,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,WAAa,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,UAAY,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,UAAY,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,UAAY,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,UAAY,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,mBAAqB,CAAC,KAAO,UAAU,CAAC,CAAC",
  "names": ["richText", "u", "x", "p", "richText1", "Link", "motion", "richText2", "richText3", "richText4", "richText5", "richText6", "richText7", "richText8", "richText9", "richText10", "richText11", "richText12", "richText13", "richText14", "richText15", "__FramerMetadata__"]
}