A challenge set for elementary-level Math Word Problems (MWP). An MWP consists of a short Natural Language narrative that describes a state of the world and poses a question about some unknown quantities.
212 PAPERS • 1 BENCHMARK
The WebQuestions dataset is a question answering dataset using Freebase as the knowledge base and contains 6,642 question-answer pairs. It was created by crawling questions through the Google Suggest API, and then obtaining answers using Amazon Mechanical Turk. The original split uses 3,778 examples for training and 2,032 for testing. All answers are defined as Freebase entities.
202 PAPERS • 3 BENCHMARKS
BEIR (Benchmarking IR) is a heterogeneous benchmark containing different information retrieval (IR) tasks. Through BEIR, it is possible to systematically study the zero-shot generalization capabilities of multiple neural retrieval approaches.
201 PAPERS • 19 BENCHMARKS
The WikiQA corpus is a publicly available set of question and sentence pairs, collected and annotated for research on open-domain question answering. In order to reflect the true information need of general users, Bing query logs were used as the question source. Each question is linked to a Wikipedia page that potentially has the answer. Because the summary section of a Wikipedia page provides the basic and usually most important information about the topic, sentences in this section were used as the candidate answers. The corpus includes 3,047 questions and 29,258 sentences, where 1,473 sentences were labeled as answer sentences to their corresponding questions.
190 PAPERS • 2 BENCHMARKS
LibriTTS is a multi-speaker English corpus of approximately 585 hours of read English speech at 24kHz sampling rate, prepared by Heiga Zen with the assistance of Google Speech and Google Brain team members. The LibriTTS corpus is designed for TTS research. It is derived from the original materials (mp3 audio files from LibriVox and text files from Project Gutenberg) of the LibriSpeech corpus. The main differences from the LibriSpeech corpus are listed below:
189 PAPERS • 1 BENCHMARK
The LAMBADA (LAnguage Modeling Broadened to Account for Discourse Aspects) benchmark is an open-ended cloze task which consists of about 10,000 passages from BooksCorpus where a missing target word is predicted in the last sentence of each passage. The missing word is constrained to always be the last word of the last sentence and there are no candidate words to choose from. Examples were filtered by humans to ensure they were possible to guess given the context, i.e., the sentences in the passage leading up to the last sentence. Examples were further filtered to ensure that missing words could not be guessed without the context, ensuring that models attempting the dataset would need to reason over the entire paragraph to answer questions.
184 PAPERS • 1 BENCHMARK
StrategyQA is a question answering benchmark where the required reasoning steps are implicit in the question, and should be inferred using a strategy. It includes 2,780 examples, each consisting of a strategy question, its decomposition, and evidence paragraphs. Questions in StrategyQA are short, topic-diverse, and cover a wide range of strategies.
178 PAPERS • 1 BENCHMARK
BC5CDR corpus consists of 1500 PubMed articles with 4409 annotated chemicals, 5818 diseases and 3116 chemical-disease interactions.
174 PAPERS • 6 BENCHMARKS
The FewRel (Few-Shot Relation Classification Dataset) contains 100 relations and 70,000 instances from Wikipedia. The dataset is divided into three subsets: training set (64 relations), validation set (16 relations) and test set (20 relations).
170 PAPERS • 4 BENCHMARKS
The Schema-Guided Dialogue (SGD) dataset consists of over 20k annotated multi-domain, task-oriented conversations between a human and a virtual assistant. These conversations involve interactions with services and APIs spanning 20 domains, ranging from banks and events to media, calendar, travel, and weather. For most of these domains, the dataset contains multiple different APIs, many of which have overlapping functionalities but different interfaces, which reflects common real-world scenarios. The wide range of available annotations can be used for intent prediction, slot filling, dialogue state tracking, policy imitation learning, language generation, user simulation learning, among other tasks in large-scale virtual assistants. Besides these, the dataset has unseen domains and services in the evaluation set to quantify the performance in zero-shot or few shot settings.
169 PAPERS • 3 BENCHMARKS
MIMIC-CXR from Massachusetts Institute of Technology presents 371,920 chest X-rays associated with 227,943 imaging studies from 65,079 patients. The studies were performed at Beth Israel Deaconess Medical Center in Boston, MA.
168 PAPERS • 2 BENCHMARKS
WMT 2016 is a collection of datasets used in shared tasks of the First Conference on Machine Translation. The conference builds on ten previous Workshops on statistical Machine Translation.
168 PAPERS • 18 BENCHMARKS
WiC is a benchmark for the evaluation of context-sensitive word embeddings. WiC is framed as a binary classification task. Each instance in WiC has a target word w, either a verb or a noun, for which two contexts are provided. Each of these contexts triggers a specific meaning of w. The task is to identify if the occurrences of w in the two contexts correspond to the same meaning or not. In fact, the dataset can also be viewed as an application of Word Sense Disambiguation in practise.
168 PAPERS • NO BENCHMARKS YET
NELL is a dataset built from the Web via an intelligent agent called Never-Ending Language Learner. This agent attempts to learn over time to read the web. NELL has accumulated over 50 million candidate beliefs by reading the web, and it is considering these at different levels of confidence. NELL has high confidence in 2,810,379 of these beliefs.
166 PAPERS • 4 BENCHMARKS
Question Answering in Context is a large-scale dataset that consists of around 14K crowdsourced Question Answering dialogs with 98K question-answer pairs in total. Data instances consist of an interactive dialog between two crowd workers: (1) a student who poses a sequence of freeform questions to learn as much as possible about a hidden Wikipedia text, and (2) a teacher who answers the questions by providing short excerpts (spans) from the text.
165 PAPERS • 1 BENCHMARK
ATOMIC is an atlas of everyday commonsense reasoning, organized through 877k textual descriptions of inferential knowledge. Compared to existing resources that center around taxonomic knowledge, ATOMIC focuses on inferential knowledge organized as typed if-then relations with variables (e.g., "if X pays Y a compliment, then Y will likely return the compliment").
162 PAPERS • NO BENCHMARKS YET
Dataset of hate speech annotated on Internet forum posts in English at sentence-level. The source forum in Stormfront, a large online community of white nacionalists. A total of 10,568 sentence have been been extracted from Stormfront and classified as conveying hate speech or not.
162 PAPERS • 1 BENCHMARK
This dataset contains 3.3K expert-level pairwise human preferences for model responses generated by 6 models in response to 80 MT-bench questions. The 6 models are GPT-4, GPT-3.5, Claud-v1, Vicuna-13B, Alpaca-13B, and LLaMA-13B. The annotators are mostly graduate students with expertise in the topic areas of each of the questions.
160 PAPERS • NO BENCHMARKS YET
PAWS-X contains 23,659 human translated PAWS evaluation pairs and 296,406 machine translated training pairs in six typologically distinct languages: French, Spanish, German, Chinese, Japanese, and Korean. All translated pairs are sourced from examples in PAWS-Wiki.
160 PAPERS • 2 BENCHMARKS
YouCook2 is the largest task-oriented, instructional video dataset in the vision community. It contains 2000 long untrimmed videos from 89 cooking recipes; on average, each distinct recipe has 22 videos. The procedure steps for each video are annotated with temporal boundaries and described by imperative English sentences (see the example below). The videos were downloaded from YouTube and are all in the third-person viewpoint. All the videos are unconstrained and can be performed by individual persons at their houses with unfixed cameras. YouCook2 contains rich recipe types and various cooking styles from all over the world.
159 PAPERS • 7 BENCHMARKS
Visual Commonsense Reasoning (VCR) is a large-scale dataset for cognition-level visual understanding. Given a challenging question about an image, machines need to present two sub-tasks: answer correctly and provide a rationale justifying its answer. The VCR dataset contains over 212K (training), 26K (validation) and 25K (testing) questions, answers and rationales derived from 110K movie scenes.
158 PAPERS • 13 BENCHMARKS
SentEval is a toolkit for evaluating the quality of universal sentence representations. SentEval encompasses a variety of tasks, including binary and multi-class classification, natural language inference and sentence similarity. The set of tasks was selected based on what appears to be the community consensus regarding the appropriate evaluations for universal sentence representations. The toolkit comes with scripts to download and preprocess datasets, and an easy interface to evaluate sentence encoders.
157 PAPERS • 2 BENCHMARKS
CMU Multimodal Opinion Sentiment and Emotion Intensity (CMU-MOSEI) is the largest dataset of sentence level sentiment analysis and emotion recognition in online videos. CMU-MOSEI contains more than 65 hours of annotated video from more than 1000 speakers and 250 topics.
154 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.
152 PAPERS • 1 BENCHMARK
The NarrativeQA dataset includes a list of documents with Wikipedia summaries, links to full stories, and questions and answers.
147 PAPERS • 1 BENCHMARK
The Cambridge Learner Corpus First Certificate in English (CLC FCE) dataset consists of short texts, written by learners of English as an additional language in response to exam prompts eliciting free-text answers and assessing mastery of the upper-intermediate proficiency level. The texts have been manually error-annotated using a taxonomy of 77 error types. The full dataset consists of 323,192 sentences. The publicly released subset of the dataset, named FCE-public, consists of 33,673 sentences split into test and training sets of 2,720 and 30,953 sentences, respectively.
146 PAPERS • 1 BENCHMARK
DocRED (Document-Level Relation Extraction Dataset) is a relation extraction dataset constructed from Wikipedia and Wikidata. Each document in the dataset is human-annotated with named entity mentions, coreference information, intra- and inter-sentence relations, and supporting evidence. DocRED requires reading multiple sentences in a document to extract entities and infer their relations by synthesizing all information of the document. Along with the human-annotated data, the dataset provides large-scale distantly supervised data.
144 PAPERS • 4 BENCHMARKS
Paraphrase Adversaries from Word Scrambling (PAWS) is a dataset contains 108,463 human-labeled and 656k noisily labeled pairs that feature the importance of modeling structure, context, and word order information for the problem of paraphrase identification. The dataset has two subsets, one based on Wikipedia and the other one based on the Quora Question Pairs (QQP) dataset.
144 PAPERS • NO BENCHMARKS YET
Visual Dialog (VisDial) dataset contains human annotated questions based on images of MS COCO dataset. This dataset was developed by pairing two subjects on Amazon Mechanical Turk to chat about an image. One person was assigned the job of a ‘questioner’ and the other person acted as an ‘answerer’. The questioner sees only the text description of an image (i.e., an image caption from MS COCO dataset) and the original image remains hidden to the questioner. Their task is to ask questions about this hidden image to “imagine the scene better”. The answerer sees the image, caption and answers the questions asked by the questioner. The two of them can continue the conversation by asking and answering questions for 10 rounds at max.
144 PAPERS • 6 BENCHMARKS
The WebNLG corpus comprises of sets of triplets describing facts (entities and relations between them) and the corresponding facts in form of natural language text. The corpus contains sets with up to 7 triplets each along with one or more reference texts for each set. The test set is split into two parts: seen, containing inputs created for entities and relations belonging to DBpedia categories that were seen in the training data, and unseen, containing inputs extracted for entities and relations belonging to 5 unseen categories.
143 PAPERS • 17 BENCHMARKS
The OLID is a hierarchical dataset to identify the type and the target of offensive texts in social media. The dataset is collected on Twitter and publicly available. There are 14,100 tweets in total, in which 13,240 are in the training set, and 860 are in the test set. For each tweet, there are three levels of labels: (A) Offensive/Not-Offensive, (B) Targeted-Insult/Untargeted, (C) Individual/Group/Other. The relationship between them is hierarchical. If a tweet is offensive, it can have a target or no target. If it is offensive to a specific target, the target can be an individual, a group, or some other objects. This dataset is used in the OffensEval-2019 competition in SemEval-2019.
142 PAPERS • 1 BENCHMARK
MultiRC (Multi-Sentence Reading Comprehension) is a dataset of short paragraphs and multi-sentence questions, i.e., questions that can be answered by combining information from multiple sentences of the paragraph. The dataset was designed with three key challenges in mind: * The number of correct answer-options for each question is not pre-specified. This removes the over-reliance on answer-options and forces them to decide on the correctness of each candidate answer independently of others. In other words, the task is not to simply identify the best answer-option, but to evaluate the correctness of each answer-option individually. * The correct answer(s) is not required to be a span in the text. * The paragraphs in the dataset have diverse provenance by being extracted from 7 different domains such as news, fiction, historical text etc., and hence are expected to be more diverse in their contents as compared to single-domain datasets. The entire corpus consists of around 10K questions
141 PAPERS • 1 BENCHMARK
Given a partial description like "she opened the hood of the car," humans can reason about the situation and anticipate what might come next ("then, she examined the engine"). SWAG (Situations With Adversarial Generations) is a large-scale dataset for this task of grounded commonsense inference, unifying natural language inference and physically grounded reasoning.
141 PAPERS • 2 BENCHMARKS
Science Question Answering (ScienceQA) is a new benchmark that consists of 21,208 multimodal multiple choice questions with diverse science topics and annotations of their answers with corresponding lectures and explanations. Out of the questions in ScienceQA, 10,332 (48.7%) have an image context, 10,220 (48.2%) have a text context, and 6,532 (30.8%) have both. Most questions are annotated with grounded lectures (83.9%) and detailed explanations (90.5%). The lecture and explanation provide general external knowledge and specific reasons, respectively, for arriving at the correct answer. To the best of our knowledge, ScienceQA is the first large-scale multimodal dataset that annotates lectures and explanations for the answers.
Wizard of Wikipedia is a large dataset with conversations directly grounded with knowledge retrieved from Wikipedia. It is used to train and evaluate dialogue systems for knowledgeable open dialogue with clear grounding
Long-range arena (LRA) is an effort toward systematic evaluation of efficient transformer models. The project aims at establishing benchmark tasks/datasets using which we can evaluate transformer-based models in a systematic way, by assessing their generalization power, computational efficiency, memory foot-print, etc. Long-Range Arena is specifically focused on evaluating model quality under long-context scenarios. The benchmark is a suite of tasks consisting of sequences ranging from 1K to 16K tokens, encompassing a wide range of data types and modalities such as text, natural, synthetic images, and mathematical expressions requiring similarity, structural, and visual-spatial reasoning.
140 PAPERS • 1 BENCHMARK
Clotho is an audio captioning dataset, consisting of 4981 audio samples, and each audio sample has five captions (a total of 24 905 captions). Audio samples are of 15 to 30 s duration and captions are eight to 20 words long.
139 PAPERS • 5 BENCHMARKS
ROCStories is a collection of commonsense short stories. The corpus consists of 100,000 five-sentence stories. Each story logically follows everyday topics created by Amazon Mechanical Turk workers. These stories contain a variety of commonsense causal and temporal relations between everyday events. Writers also develop an additional 3,742 Story Cloze Test stories which contain a four-sentence-long body and two candidate endings. The endings were collected by asking Mechanical Turk workers to write both a right ending and a wrong ending after eliminating original endings of given short stories. Both endings were required to make logical sense and include at least one character from the main story line. The published ROCStories dataset is constructed with ROCStories as a training set that includes 98,162 stories that exclude candidate wrong endings, an evaluation set, and a test set, which have the same structure (1 body + 2 candidate endings) and a size of 1,871.
139 PAPERS • 2 BENCHMARKS
This paper introduces the pipeline to scale the largest dataset in egocentric vision EPIC-KITCHENS. The effort culminates in EPIC-KITCHENS-100, a collection of 100 hours, 20M frames, 90K actions in 700 variable-length videos, capturing long-term unscripted activities in 45 environments, using head-mounted cameras. Compared to its previous version (EPIC-KITCHENS-55), EPIC-KITCHENS-100 has been annotated using a novel pipeline that allows denser (54% more actions per minute) and more complete annotations of fine-grained actions (+128% more action segments). This collection also enables evaluating the "test of time" - i.e. whether models trained on data collected in 2018 can generalise to new footage collected under the same hypotheses albeit "two years on". The dataset is aligned with 6 challenges: action recognition (full and weak supervision), action detection, action anticipation, cross-modal retrieval (from captions), as well as unsupervised domain adaptation for action recognition.
136 PAPERS • 7 BENCHMARKS
The One Billion Word dataset is a dataset for language modeling. The training/held-out data was produced from the WMT 2011 News Crawl data using a combination of Bash shell and Perl scripts.
134 PAPERS • 2 BENCHMARKS
At the end of 2017 the Civil Comments platform shut down and chose make their ~2m public comments from their platform available in a lasting open archive so that researchers could understand and improve civility in online conversations for years to come. Jigsaw sponsored this effort and extended annotation of this data by human raters for various toxic conversational attributes.
131 PAPERS • 1 BENCHMARK
MIcrosoft News Dataset (MIND) is a large-scale dataset for news recommendation research. It was collected from anonymized behavior logs of Microsoft News website. The mission of MIND is to serve as a benchmark dataset for news recommendation and facilitate the research in news recommendation and recommender systems area.
127 PAPERS • 1 BENCHMARK
SearchQA was built using an in-production, commercial search engine. It closely reflects the full pipeline of a (hypothetical) general question-answering system, which consists of information retrieval and answer synthesis.
126 PAPERS • 1 BENCHMARK
A corpus of parallel text in 21 European languages from the proceedings of the European Parliament.
125 PAPERS • NO BENCHMARKS YET
The BLUE benchmark consists of five different biomedicine text-mining tasks with ten corpora. These tasks cover a diverse range of text genres (biomedical literature and clinical notes), dataset sizes, and degrees of difficulty and, more importantly, highlight common biomedicine text-mining challenges.
122 PAPERS • NO BENCHMARKS YET
e-SNLI is used for various goals, such as obtaining full sentence justifications of a model's decisions, improving universal sentence representations and transferring to out-of-domain NLI datasets.
122 PAPERS • 1 BENCHMARK
The Yahoo! Answers topic classification dataset is constructed using 10 largest main categories. Each class contains 140,000 training samples and 6,000 testing samples. Therefore, the total number of training samples is 1,400,000 and testing samples 60,000 in this dataset. From all the answers and other meta-information, we only used the best answer content and the main category information. Source:github
121 PAPERS • 2 BENCHMARKS
SciERC dataset is a collection of 500 scientific abstract annotated with scientific entities, their relations, and coreference clusters. The abstracts are taken from 12 AI conference/workshop proceedings in four AI communities, from the Semantic Scholar Corpus. SciERC extends previous datasets in scientific articles SemEval 2017 Task 10 and SemEval 2018 Task 7 by extending entity types, relation types, relation coverage, and adding cross-sentence relations using coreference links.
118 PAPERS • 7 BENCHMARKS