General Language Understanding Evaluation (GLUE) benchmark is a collection of nine natural language understanding tasks, including single-sentence tasks CoLA and SST-2, similarity and paraphrasing tasks MRPC, STS-B and QQP, and natural language inference tasks MNLI, QNLI, RTE and WNLI.
2,647 PAPERS • 45 BENCHMARKS
The Stanford Sentiment Treebank is a corpus with fully labeled parse trees that allows for a complete analysis of the compositional effects of sentiment in language. The corpus is based on the dataset introduced by Pang and Lee (2005) and consists of 11,855 single sentences extracted from movie reviews. It was parsed with the Stanford parser and includes a total of 215,154 unique phrases from those parse trees, each annotated by 3 human judges.
1,989 PAPERS • 13 BENCHMARKS
The Multi-Genre Natural Language Inference (MultiNLI) dataset has 433K sentence pairs. Its size and mode of collection are modeled closely like SNLI. MultiNLI offers ten distinct genres (Face-to-face, Telephone, 9/11, Travel, Letters, Oxford University Press, Slate, Verbatim, Goverment and Fiction) of written and spoken English data. There are matched dev/test sets which are derived from the same sources as those in the training set, and mismatched sets which do not closely resemble any seen at training time.
1,637 PAPERS • 8 BENCHMARKS
The QNLI (Question-answering NLI) dataset is a Natural Language Inference dataset automatically derived from the Stanford Question Answering Dataset v1.1 (SQuAD). SQuAD v1.1 consists of question-paragraph pairs, where one of the sentences in the paragraph (drawn from Wikipedia) contains the answer to the corresponding question (written by an annotator). The dataset was converted into sentence pair classification by forming a pair between each question and each sentence in the corresponding context, and filtering out pairs with low lexical overlap between the question and the context sentence. The task is to determine whether the context sentence contains the answer to the question. This modified version of the original task removes the requirement that the model select the exact answer, but also removes the simplifying assumptions that the answer is always present in the input and that lexical overlap is a reliable cue. The QNLI dataset is part of GLUE benchmark.
1,028 PAPERS • 5 BENCHMARKS
The WikiText language modeling dataset is a collection of over 100 million tokens extracted from the set of verified Good and Featured articles on Wikipedia. The dataset is available under the Creative Commons Attribution-ShareAlike License.
794 PAPERS • 5 BENCHMARKS
Microsoft Research Paraphrase Corpus (MRPC) is a corpus consists of 5,801 sentence pairs collected from newswire articles. Each pair is labelled if it is a paraphrase or not by human annotators. The whole set is divided into a training subset (4,076 sentence pairs of which 2,753 are paraphrases) and a test subset (1,725 pairs of which 1,147 are paraphrases).
682 PAPERS • 7 BENCHMARKS
MMLU (Massive Multitask Language Understanding) is a new benchmark designed to measure knowledge acquired during pretraining by evaluating models exclusively in zero-shot and few-shot settings. This makes the benchmark more challenging and more similar to how we evaluate humans. The benchmark covers 57 subjects across STEM, the humanities, the social sciences, and more. It ranges in difficulty from an elementary level to an advanced professional level, and it tests both world knowledge and problem solving ability. Subjects range from traditional areas, such as mathematics and history, to more specialized areas like law and ethics. The granularity and breadth of the subjects makes the benchmark ideal for identifying a model’s blind spots.
614 PAPERS • 26 BENCHMARKS
TriviaQA is a realistic text-based question answering dataset which includes 950K question-answer pairs from 662K documents collected from Wikipedia and the web. This dataset is more challenging than standard QA benchmark datasets such as Stanford Question Answering Dataset (SQuAD), as the answers for a question may not be directly obtained by span prediction and the context is very long. TriviaQA dataset consists of both human-verified and machine-generated QA subsets.
600 PAPERS • 5 BENCHMARKS
486 PAPERS • 2 BENCHMARKS
CNN/Daily Mail is a dataset for text summarization. Human generated abstractive summary bullets were generated from news stories in CNN and Daily Mail websites as questions (with one of the entities hidden), and stories as the corresponding passages from which the system is expected to answer the fill-in the-blank question. The authors released the scripts that crawl, extract and generate pairs of passages and questions from these websites.
459 PAPERS • 13 BENCHMARKS
BoolQ is a question answering dataset for yes/no questions containing 15942 examples. These questions are naturally occurring – they are generated in unprompted and unconstrained settings. Each example is a triplet of (question, passage, answer), with the title of the page as optional additional context.
370 PAPERS • 2 BENCHMARKS
DailyDialog is a high-quality multi-turn open-domain English dialog dataset. It contains 13,118 dialogues split into a training set with 11,118 dialogues and validation and test sets with 1000 dialogues each. On average there are around 8 speaker turns per dialogue with around 15 tokens per turn.
363 PAPERS • 2 BENCHMARKS
The ReAding Comprehension dataset from Examinations (RACE) dataset is a machine reading comprehension dataset consisting of 27,933 passages and 97,867 questions from English exams, targeting Chinese students aged 12-18. RACE consists of two subsets, RACE-M and RACE-H, from middle school and high school exams, respectively. RACE-M has 28,293 questions and RACE-H has 69,574. Each question is associated with 4 candidate answers, one of which is correct. The data generation process of RACE differs from most machine reading comprehension datasets - instead of generating questions and answers by heuristics or crowd-sourcing, questions in RACE are specifically designed for testing human reading skills, and are created by domain experts.
345 PAPERS • 4 BENCHMARKS
BookCorpus is a large collection of free novel books written by unpublished authors, which contains 11,038 books (around 74M sentences and 1G words) of 16 different sub-genres (e.g., Romance, Historical, Adventure, etc.).
318 PAPERS • NO BENCHMARKS YET
The Multi-domain Wizard-of-Oz (MultiWOZ) dataset is a large-scale human-human conversational corpus spanning over seven domains, containing 8438 multi-turn dialogues, with each dialogue averaging 14 turns. Different from existing standard datasets like WOZ and DSTC2, which contain less than 10 slots and only a few hundred values, MultiWOZ has 30 (domain, slot) pairs and over 4,500 possible values. The dialogues span seven domains: restaurant, hotel, attraction, taxi, train, hospital and police.
314 PAPERS • 15 BENCHMARKS
The Winograd Schema Challenge was introduced both as an alternative to the Turing Test and as a test of a system’s ability to do commonsense reasoning. A Winograd schema is a pair of sentences differing in one or two words with a highly ambiguous pronoun, resolved differently in the two sentences, that appears to require commonsense knowledge to be resolved correctly. The examples were designed to be easily solvable by humans but difficult for machines, in principle requiring a deep understanding of the content of the text and the situation it describes.
302 PAPERS • 2 BENCHMARKS
The Choice Of Plausible Alternatives (COPA) evaluation provides researchers with a tool for assessing progress in open-domain commonsense causal reasoning. COPA consists of 1000 questions, split equally into development and test sets of 500 questions each. Each question is composed of a premise and two alternatives, where the task is to select the alternative that more plausibly has a causal relation with the premise. The correct alternative is randomized so that the expected performance of randomly guessing is 50%.
268 PAPERS • 2 BENCHMARKS
OpenBookQA is a new kind of question-answering dataset modeled after open book exams for assessing human understanding of a subject. It consists of 5,957 multiple-choice elementary-level science questions (4,957 train, 500 dev, 500 test), which probe the understanding of a small “book” of 1,326 core science facts and the application of these facts to novel situations. For training, the dataset includes a mapping from each question to the core science fact it was designed to probe. Answering OpenBookQA questions requires additional broad common knowledge, not contained in the book. The questions, by design, are answered incorrectly by both a retrieval-based algorithm and a word co-occurrence algorithm. Additionally, the dataset includes a collection of 5,167 crowd-sourced common knowledge facts, and an expanded version of the train/dev/test questions where each question is associated with its originating core fact, a human accuracy score, a clarity score, and an anonymized crowd-worker
222 PAPERS • 2 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.
166 PAPERS • 1 BENCHMARK
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.
165 PAPERS • 6 BENCHMARKS
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
140 PAPERS • 2 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.
136 PAPERS • 2 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
ReDial (Recommendation Dialogues) is an annotated dataset of dialogues, where users recommend movies to each other. The dataset consists of over 10,000 conversations centered around the theme of providing movie recommendations.
85 PAPERS • 2 BENCHMARKS
VQG is a collection of datasets for visual question generation. VQG questions were collected by crowdsourcing the task on Amazon Mechanical Turk (AMT). The authors provided details on the prompt and the specific instructions for all the crowdsourcing tasks in this paper in the supplementary material. The prompt was successful at capturing nonliteral questions. Images were taken from the MSCOCO dataset.
77 PAPERS • 1 BENCHMARK
End-to-End NLG Challenge (E2E) aims to assess whether recent end-to-end NLG systems can generate more complex output by learning from datasets containing higher lexical richness, syntactic complexity and diverse discourse phenomena.
67 PAPERS • 4 BENCHMARKS
LogiQA consists of 8,678 QA instances, covering multiple types of deductive reasoning. Results show that state-of-the-art neural models perform by far worse than human ceiling. The dataset can also serve as a benchmark for reinvestigating logical AI under the deep learning NLP setting.
63 PAPERS • 1 BENCHMARK
Sentence Compression is a dataset where the syntactic trees of the compressions are subtrees of their uncompressed counterparts, and hence where supervised systems which require a structural alignment between the input and output can be successfully trained.
61 PAPERS • 1 BENCHMARK
Quora Question Pairs (QQP) dataset consists of over 400,000 question pairs, and each question pair is annotated with a binary value indicating whether the two questions are paraphrase of each other.
53 PAPERS • 10 BENCHMARKS
The Machine Translation of Noisy Text (MTNT) dataset is a Machine Translation dataset that consists of noisy comments on Reddit and professionally sourced translation. The translation are between French, Japanese and French, with between 7k and 37k sentence per language pair.
51 PAPERS • NO BENCHMARKS YET
MuTual is a retrieval-based dataset for multi-turn dialogue reasoning, which is modified from Chinese high school English listening comprehension test data. It tests dialogue reasoning via next utterance prediction.
50 PAPERS • NO BENCHMARKS YET
The Image Paragraph Captioning dataset allows researchers to benchmark their progress in generating paragraphs that tell a story about an image. The dataset contains 19,561 images from the Visual Genome dataset. Each image contains one paragraph. The training/val/test sets contains 14,575/2,487/2,489 images.
29 PAPERS • 1 BENCHMARK
The AMR Bank is a set of English sentences paired with simple, readable semantic representations. Version 3.0 released in 2020 consists of 59,255 sentences.
22 PAPERS • 1 BENCHMARK
COunter NArratives through Nichesourcing (CONAN) is a dataset that consists of 4,078 pairs over the 3 languages. Additionally, 3 types of metadata are provided: expert demographics, hate speech sub-topic and counter-narrative type. The dataset is augmented through translation (from Italian/French to English) and paraphrasing, which brought the total number of pairs to 14.988.
21 PAPERS • NO BENCHMARKS YET
Moral Stories is a crowd-sourced dataset of structured narratives that describe normative and norm-divergent actions taken by individuals to accomplish certain intentions in concrete situations, and their respective consequences.
19 PAPERS • NO BENCHMARKS YET
Opusparcus is a paraphrase corpus for six European languages: German, English, Finnish, French, Russian, and Swedish. The paraphrases are extracted from the OpenSubtitles2016 corpus, which contains subtitles from movies and TV shows.
15 PAPERS • NO BENCHMARKS YET
The PROST (Physical Reasoning about Objects Through Space and Time) dataset contains 18,736 multiple-choice questions made from 14 manually curated templates, covering 10 physical reasoning concepts. All questions are designed to probe both causal and masked language models in a zero-shot setting.
9 PAPERS • 1 BENCHMARK
Logic2Text is a large-scale dataset with 10,753 descriptions involving common logic types paired with the underlying logical forms. The logical forms show diversified graph structure of free schema, which poses great challenges on the model's ability to understand the semantics.
8 PAPERS • NO BENCHMARKS YET
The VNHSGE (VietNamese High School Graduation Examination) dataset, developed exclusively for evaluating large language models (LLMs), is introduced in this article. The dataset, which covers nine subjects, was generated from the Vietnamese National High School Graduation Examination and comparable tests. 300 literary essays have been included, and there are over 19,000 multiple-choice questions on a range of topics. The dataset assesses LLMs in multitasking situations such as question answering, text generation, reading comprehension, visual question answering, and more by including both textual data and accompanying images. Using ChatGPT and BingChat, we evaluated LLMs on the VNHSGE dataset and contrasted their performance with that of Vietnamese students to see how well they performed. The results show that ChatGPT and BingChat both perform at a human level in a number of areas, including literature, English, history, geography, and civics education. They still have space to grow, t
7 PAPERS • 9 BENCHMARKS
Paper | Github | Dataset| Model
5 PAPERS • 1 BENCHMARK
A new English-French test set for the evaluation of Machine Translation (MT) for informal, written bilingual dialogue. The test set contains 144 spontaneous dialogues (5,700+ sentences) between native English and French speakers, mediated by one of two neural MT systems in a range of role-play settings. The dialogues are accompanied by fine-grained sentence-level judgments of MT quality, produced by the dialogue participants themselves, as well as by manually normalised versions and reference translations produced a posteriori.
4 PAPERS • 1 BENCHMARK
Benchmark dataset for abstracts and titles of 100,000 ArXiv scientific papers. This dataset contains 10 classes and is balanced (exactly 10,000 per class). The classes include subcategories of computer science, physics, and math.
ChatHaruhi is a dataset covering 32 Chinese / English TV / anime characters with over 54k simulated dialogues.
3 PAPERS • NO BENCHMARKS YET
Goal is a novel dataset of football (or 'soccer') highlights videos with transcribed live commentaries in English. As the course of a game is unpredictable, so are commentaries, which makes them a unique resource to investigate dynamic language grounding.
Dataset Summary CANNOT is a dataset that focuses on negated textual pairs. It currently contains 77,376 samples, of which roughly of them are negated pairs of sentences, and the other half are not (they are paraphrased versions of each other).
2 PAPERS • NO BENCHMARKS YET
The QTUNA dataset is the result of a series of elicitation experiments in which human speakers were asked to perform a linguistic task that invites the use of quantified expressions in order to inform possible Natural Language Generation algorithms that mimic humans' use of quantified expressions.
TextBox 2.0 is a comprehensive and unified library for text generation, focusing on the use of pre-trained language models (PLMs). The library covers 13 common text generation tasks and their corresponding 83 datasets and further incorporates 45 PLMs covering general, translation, Chinese, dialogue, controllable, distilled, prompting, and lightweight PLMs.
YTD-18M is a large-scale corpus of 18M video-based dialogues, constructed from web videos: crucial to the data collection pipeline is a pretrained language model that converts error-prone automatic transcripts to a cleaner dialogue format while maintaining meaning.