The Stanford Question Answering Dataset (SQuAD) is a collection of question-answer pairs derived from Wikipedia articles. In SQuAD, the correct answers of questions can be any sequence of tokens in the given text. Because the questions and answers are produced by humans through crowdsourcing, it is more diverse than some other question-answering datasets. SQuAD 1.1 contains 107,785 question-answer pairs on 536 articles. SQuAD2.0 (open-domain SQuAD, SQuAD-Open), the latest version, combines the 100,000 questions in SQuAD1.1 with over 50,000 un-answerable questions written adversarially by crowdworkers in forms that are similar to the answerable ones.

1,649 PAPERS • 13 BENCHMARKS

The Universal Dependencies (UD) project seeks to develop cross-linguistically consistent treebank annotation of morphology and syntax for multiple languages. The first version of the dataset was released in 2015 and consisted of 10 treebanks over 10 languages. Version 2.7 released in 2020 consists of 183 treebanks over 104 languages. The annotation consists of UPOS (universal part-of-speech tags), XPOS (language-specific part-of-speech tags), Feats (universal morphological features), Lemmas, dependency heads and universal dependency labels.

480 PAPERS • 15 BENCHMARKS

The Cross-lingual Natural Language Inference (XNLI) corpus is the extension of the Multi-Genre NLI (MultiNLI) corpus to 15 languages. The dataset was created by manually translating the validation and test sets of MultiNLI into each of those 15 languages. The English training set was machine translated for all languages. The dataset is composed of 122k train, 2490 validation and 5010 test examples.

283 PAPERS • 11 BENCHMARKS

Common Voice is an audio dataset that consists of a unique MP3 and corresponding text file. There are 9,283 recorded hours in the dataset. The dataset also includes demographic metadata like age, sex, and accent. The dataset consists of 7,335 validated hours in 60 languages.

215 PAPERS • 262 BENCHMARKS

OpenSubtitles is collection of multilingual parallel corpora. The dataset is compiled from a large database of movie and TV subtitles and includes a total of 1689 bitexts spanning 2.6 billion sentences across 60 languages.

189 PAPERS • 1 BENCHMARK

MuST-C currently represents the largest publicly available multilingual corpus (one-to-many) for speech translation. It covers eight language directions, from English to German, Spanish, French, Italian, Dutch, Portuguese, Romanian and Russian. The corpus consists of audio, transcriptions and translations of English TED talks, and it comes with a predefined training, validation and test split.

169 PAPERS • 2 BENCHMARKS

XQuAD (Cross-lingual Question Answering Dataset) is a benchmark dataset for evaluating cross-lingual question answering performance. The dataset consists of a subset of 240 paragraphs and 1190 question-answer pairs from the development set of SQuAD v1.1 (Rajpurkar et al., 2016) together with their professional translations into ten languages: Spanish, German, Greek, Russian, Turkish, Arabic, Vietnamese, Thai, Chinese, and Hindi. Consequently, the dataset is entirely parallel across 11 languages.

133 PAPERS • 2 BENCHMARKS

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance. MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic, German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between 4 different languages on average.

131 PAPERS • 1 BENCHMARK

This corpus comprises of monolingual data for 100+ languages and also includes data for romanized languages. This was constructed using the urls and paragraph indices provided by the CC-Net repository by processing January-December 2018 Commoncrawl snapshots. Each file comprises of documents separated by double-newlines and paragraphs within the same document separated by a newline. The data is generated using the open source CC-Net repository.

71 PAPERS • NO BENCHMARKS YET

OntoNotes 5.0 is a large corpus comprising various genres of text (news, conversational telephone speech, weblogs, usenet newsgroups, broadcast, talk shows) in three languages (English, Chinese, and Arabic) with structural information (syntax and predicate argument structure) and shallow semantics (word sense linked to an ontology and coreference).

65 PAPERS • 12 BENCHMARKS

OSCAR or Open Super-large Crawled ALMAnaCH coRpus is a huge multilingual corpus obtained by language classification and filtering of the Common Crawl corpus using the goclassy architecture. The dataset used for training multilingual models such as BART incorporates 138 GB of text.

42 PAPERS • NO BENCHMARKS YET

WikiLingua includes ~770k article and summary pairs in 18 languages from WikiHow. Gold-standard article-summary alignments across languages are extracted by aligning the images that are used to describe each how-to step in an article.

40 PAPERS • 5 BENCHMARKS

WikiAnn is a dataset for cross-lingual name tagging and linking based on Wikipedia articles in 295 languages.

36 PAPERS • 7 BENCHMARKS

Multilingual Knowledge Questions and Answers (MKQA) is an open-domain question answering evaluation set comprising 10k question-answer pairs aligned across 26 typologically diverse languages (260k question-answer pairs in total). The goal of this dataset is to provide a challenging benchmark for question answering quality across a wide set of languages. Answers are based on a language-independent data representation, making results comparable across languages and independent of language-specific passages. With 26 languages, this dataset supplies the widest range of languages to-date for evaluating question answering.

30 PAPERS • NO BENCHMARKS YET

Arabic Sentiment Tweets Dataset (ASTD) is an Arabic social sentiment analysis dataset gathered from Twitter. It consists of about 10,000 tweets which are classified as objective, subjective positive, subjective negative, and subjective mixed.

28 PAPERS • 1 BENCHMARK

CoVoST is a large-scale multilingual speech-to-text translation corpus. Its latest 2nd version covers translations from 21 languages into English and from English into 15 languages. It has total 2880 hours of speech and is diversified with 78K speakers and 66 accents.

26 PAPERS • NO BENCHMARKS YET

AVSpeech is a large-scale audio-visual dataset comprising speech clips with no interfering background signals. The segments are of varying length, between 3 and 10 seconds long, and in each clip the only visible face in the video and audible sound in the soundtrack belong to a single speaking person. In total, the dataset contains roughly 4700 hours of video segments with approximately 150,000 distinct speakers, spanning a wide variety of people, languages and face poses.

25 PAPERS • NO BENCHMARKS YET

XL-Sum is a comprehensive and diverse dataset for abstractive summarization comprising 1 million professionally annotated article-summary pairs from BBC, extracted using a set of carefully designed heuristics. The dataset covers 44 languages ranging from low to high-resource, for many of which no public dataset is currently available. XL-Sum is highly abstractive, concise, and of high quality, as indicated by human and intrinsic evaluation.

23 PAPERS • NO BENCHMARKS YET

iSarcasmEval is the first shared task to target intended sarcasm detection: the data for this task was provided and labelled by the authors of the texts themselves. Such an approach minimises the downfalls of other methods to collect sarcasm data, which rely on distant supervision or third-party annotations. The shared task contains two languages, English and Arabic, and three subtasks: sarcasm detection, sarcasm category classification, and pairwise sarcasm identification given a sarcastic sentence and its non-sarcastic rephrase. The task received submissions from 60 different teams, with the sarcasm detection task being the most popular. Most of the participating teams utilised pre-trained language models. In this paper, we provide an overview of the task, data, and participating teams.

19 PAPERS • NO BENCHMARKS YET

The first parallel corpus composed from United Nations documents published by the original data creator. The parallel corpus presented consists of manually translated UN documents from the last 25 years (1990 to 2014) for the six official UN languages, Arabic, Chinese, English, French, Russian, and Spanish.

16 PAPERS • NO BENCHMARKS YET

LABR is a large sentiment analysis dataset to-date for the Arabic language. It consists of over 63,000 book reviews, each rated on a scale of 1 to 5 stars.

15 PAPERS • 1 BENCHMARK

OntoNotes Release 4.0 contains the content of earlier releases -- OntoNotes Release 1.0 LDC2007T21, OntoNotes Release 2.0 LDC2008T04 and OntoNotes Release 3.0 LDC2009T24 -- and adds newswire, broadcast news, broadcast conversation and web data in English and Chinese and newswire data in Arabic. This cumulative publication consists of 2.4 million words as follows: 300k words of Arabic newswire 250k words of Chinese newswire, 250k words of Chinese broadcast news, 150k words of Chinese broadcast conversation and 150k words of Chinese web text and 600k words of English newswire, 200k word of English broadcast news, 200k words of English broadcast conversation and 300k words of English web text.

15 PAPERS • 1 BENCHMARK

ArSarcasm-v2 is an extension of the original ArSarcasm dataset published along with the paper From Arabic Sentiment Analysis to Sarcasm Detection: The ArSarcasm Dataset. ArSarcasm-v2 conisists of ArSarcasm along with portions of DAICT corpus and some new tweets. Each tweet was annotated for sarcasm, sentiment and dialect. The final dataset consists of 15,548 tweets divided into 12,548 training tweets and 3,000 testing tweets. ArSarcasm-v2 was used and released as a part of the shared task on sarcasm detection and sentiment analysis in Arabic.

14 PAPERS • NO BENCHMARKS YET

The Image-Grounded Language Understanding Evaluation (IGLUE) benchmark brings together—by both aggregating pre-existing datasets and creating new ones—visual question answering, cross-modal retrieval, grounded reasoning, and grounded entailment tasks across 20 diverse languages. The benchmark enables the evaluation of multilingual multimodal models for transfer learning, not only in a zero-shot setting, but also in newly defined few-shot learning setups.

14 PAPERS • 13 BENCHMARKS

XGLUE is an evaluation benchmark XGLUE,which is composed of 11 tasks that span 19 languages. For each task, the training data is only available in English. This means that to succeed at XGLUE, a model must have a strong zero-shot cross-lingual transfer capability to learn from the English data of a specific task and transfer what it learned to other languages. Comparing to its concurrent work XTREME, XGLUE has two characteristics: First, it includes cross-lingual NLU and cross-lingual NLG tasks at the same time; Second, besides including 5 existing cross-lingual tasks (i.e. NER, POS, MLQA, PAWS-X and XNLI), XGLUE selects 6 new tasks from Bing scenarios as well, including News Classification (NC), Query-Ad Matching (QADSM), Web Page Ranking (WPR), QA Matching (QAM), Question Generation (QG) and News Title Generation (NTG). Such diversities of languages, tasks and task origin provide a comprehensive benchmark for quantifying the quality of a pre-trained model on cross-lingual natural lan

14 PAPERS • 3 BENCHMARKS

ArSarcasm is a new Arabic sarcasm detection dataset. The dataset was created using previously available Arabic sentiment analysis datasets (SemEval 2017 and ASTD) and adds sarcasm and dialect labels to them. The dataset contains 10,547 tweets, 1,682 (16%) of which are sarcastic.

11 PAPERS • NO BENCHMARKS YET

Extracted from the Tashkeela Corpus, the dataset consists of 55K lines containing about 2.3M words.

11 PAPERS • 1 BENCHMARK

xSID, a new evaluation benchmark for cross-lingual (X) Slot and Intent Detection in 13 languages from 6 language families, including a very low-resource dialect, covering Arabic (ar), Chinese (zh), Danish (da), Dutch (nl), English (en), German (de), Indonesian (id), Italian (it), Japanese (ja), Kazakh (kk), Serbian (sr), Turkish (tr) and an Austro-Bavarian German dialect, South Tyrolean (de-st).

10 PAPERS • NO BENCHMARKS YET

CVSS is a massively multilingual-to-English speech to speech translation (S2ST) corpus, covering sentence-level parallel S2ST pairs from 21 languages into English. CVSS is derived from the Common Voice speech corpus and the CoVoST 2 speech-to-text translation (ST) corpus, by synthesizing the translation text from CoVoST 2 into speech using state-of-the-art TTS systems

9 PAPERS • NO BENCHMARKS YET

X-FACT is a large publicly available multilingual dataset for factual verification of naturally existing real-world claims. The dataset contains short statements in 25 languages and is labeled for veracity by expert fact-checkers. The dataset includes a multilingual evaluation benchmark that measures both out-of-domain generalization, and zero-shot capabilities of the multilingual models.

9 PAPERS • 1 BENCHMARK

Synbols is a dataset generator designed for probing the behavior of learning algorithms. By defining the distribution over latent factors one can craft a dataset specifically tailored to answer specific questions about a given algorithm.

8 PAPERS • NO BENCHMARKS YET

Global Voices is a multilingual dataset for evaluating cross-lingual summarization methods. It is extracted from social-network descriptions of Global Voices news articles to cheaply collect evaluation data for into-English and from-English summarization in 15 languages.

7 PAPERS • NO BENCHMARKS YET

ArCOV-19 is an Arabic COVID-19 Twitter dataset that covers the period from 27th of January till 30th of April 2020. ArCOV-19 is the first publicly-available Arabic Twitter dataset covering COVID-19 pandemic that includes over 1M tweets alongside the propagation networks of the most-popular subset of them (i.e., most-retweeted and -liked).

5 PAPERS • NO BENCHMARKS YET

A large-scale corpus of Gulf Arabic consisting of 110 million words from 1,200 forum novels.

5 PAPERS • NO BENCHMARKS YET

MuMiN is a misinformation graph dataset containing rich social media data (tweets, replies, users, images, articles, hashtags), spanning 21 million tweets belonging to 26 thousand Twitter threads, each of which have been semantically linked to 13 thousand fact-checked claims across dozens of topics, events and domains, in 41 different languages, spanning more than a decade.

4 PAPERS • 3 BENCHMARKS

An English-Arabic named entity transliteration and classification dataset built from freely available parallel translation corpora. The dataset contains 79,924 instances, each instance is a triplet (e, a, c), where e is the English named entity, a is its Arabic transliteration and c is its class that can be either a Person, a Location, or an Organization. The ANETAC dataset is mainly aimed for the researchers that are working on Arabic named entity transliteration, but it can also be used for named entity classification purposes.

3 PAPERS • NO BENCHMARKS YET

The DAWT dataset consists of Densely Annotated Wikipedia Texts across multiple languages. The annotations include labeled text mentions mapping to entities (represented by their Freebase machine ids) as well as the type of the entity. The data set contains total of 13.6M articles, 5.0B tokens, 13.8M mention entity co-occurrences. DAWT contains 4.8 times more anchor text to entity links than originally present in the Wikipedia markup. Moreover, it spans several languages including English, Spanish, Italian, German, French and Arabic.

3 PAPERS • NO BENCHMARKS YET

GeoCoV19 is a large-scale Twitter dataset containing more than 524 million multilingual tweets. The dataset contains around 378K geotagged tweets and 5.4 million tweets with Place information. The annotations include toponyms from the user location field and tweet content and resolve them to geolocations such as country, state, or city level. In this case, 297 million tweets are annotated with geolocation using the user location field and 452 million tweets using tweet content.

3 PAPERS • NO BENCHMARKS YET

POST-COMPETITION: The official competition is now over, but you are welcome to develop and test new solutions on this website. All data with gold labels (training, developing, and test) are available here. The test data in this archive do not include the instances from the Equity Evaluation Corpus (EEC) used for bias evaluation. The EEC corpus is available here.

3 PAPERS • NO BENCHMARKS YET

AM2iCo is a wide-coverage and carefully designed cross-lingual and multilingual evaluation set. It aims to assess the ability of state-of-the-art representation models to reason over cross-lingual lexical-level concept alignment in context for 14 language pairs.

2 PAPERS • NO BENCHMARKS YET

ArCOV19-Rumors is an Arabic COVID-19 Twitter dataset for misinformation detection composed of tweets containing claims from 27th January till the end of April 2020.

2 PAPERS • NO BENCHMARKS YET

AraCOVID19-MFH is a manually annotated multi-label Arabic COVID-19 fake news and hate speech detection dataset. The dataset contains 10,828 Arabic tweets annotated with 10 different labels.

2 PAPERS • NO BENCHMARKS YET

MINTAKA is a complex, natural, and multilingual dataset designed for experimenting with end-to-end question-answering models. It is composed of 20,000 question-answer pairs collected in English, annotated with Wikidata entities, and translated into Arabic, French, German, Hindi, Italian, Japanese, Portuguese, and Spanish for a total of 180,000 samples. Mintaka includes 8 types of complex questions, including superlative, intersection, and multi-hop questions, which were naturally elicited from crowd workers.

2 PAPERS • NO BENCHMARKS YET

NSURL-2019 Shared Task 8: Semantic Question Similarity in Arabic

2 PAPERS • NO BENCHMARKS YET

Stanceosaurus is a corpus of 28,033 tweets in English, Hindi, and Arabic annotated with stance towards 251 misinformation claims. The claims in Stanceosaurus originate from 15 fact-checking sources that cover diverse geographical regions and cultures. Unlike existing stance datasets, it introduces a more fine-grained 5-class labeling strategy with additional subcategories to distinguish implicit stance.

2 PAPERS • NO BENCHMARKS YET

A new text effects dataset with 141,081 text effect/glyph pairs in total. The dataset consists of 152 professionally designed text effects rendered on glyphs, including English letters, Chinese characters, and Arabic numerals.

2 PAPERS • NO BENCHMARKS YET

Arabic Multi Fonts Dataset A multi-word multi-font Arabic word-image dataset.

1 PAPER • NO BENCHMARKS YET

The Archive Query Log (AQL) is a previously unused, comprehensive query log collected at the Internet Archive over the last 25 years. Its first version includes 356 million queries, 166 million search result pages, and 1.7 billion search results across 550 search providers. Although many query logs have been studied in the literature, the search providers that own them generally do not publish their logs to protect user privacy and vital business data. The AQL is the first publicly available query log that combines size, scope, and diversity, enabling research on new retrieval models and search engine analyses. Provided in a privacy-preserving manner, it promotes open research as well as more transparency and accountability in the search industry.

1 PAPER • NO BENCHMARKS YET