The Question Answering by Search And Reading (QUASAR) is a large-scale dataset consisting of QUASAR-S and QUASAR-T. Each of these datasets is built to focus on evaluating systems devised to understand a natural language query, a large corpus of texts and to extract an answer to the question from the corpus. Specifically, QUASAR-S comprises 37,012 fill-in-the-gaps questions that are collected from the popular website Stack Overflow using entity tags. The QUASAR-T dataset contains 43,012 open-domain questions collected from various internet sources. The candidate documents for each question in this dataset are retrieved from an Apache Lucene based search engine built on top of the ClueWeb09 dataset.
45 PAPERS • 1 BENCHMARK
QReCC contains 14K conversations with 81K question-answer pairs. QReCC is built on questions from TREC CAsT, QuAC and Google Natural Questions. While TREC CAsT and QuAC datasets contain multi-turn conversations, Natural Questions is not a conversational dataset. Questions in NQ dataset were used as prompts to create conversations explicitly balancing types of context-dependent questions, such as anaphora (co-references) and ellipsis.
44 PAPERS • NO BENCHMARKS YET
CoS-E consists of human explanations for commonsense reasoning in the form of natural language sequences and highlighted annotations
43 PAPERS • NO BENCHMARKS YET
DuoRC contains 186,089 unique question-answer pairs created from a collection of 7680 pairs of movie plots where each pair in the collection reflects two versions of the same movie.
42 PAPERS • 1 BENCHMARK
WikiMovies is a dataset for question answering for movies content. It contains ~100k questions in the movie domain, and was designed to be answerable by using either a perfect KB (based on OMDb),
39 PAPERS • NO BENCHMARKS YET
InsuranceQA is a question answering dataset for the insurance domain, the data stemming from the website Insurance Library. There are 12,889 questions and 21,325 answers in the training set. There are 2,000 questions and 3,354 answers in the validation set. There are 2,000 questions and 3,308 answers in the test set.
38 PAPERS • NO BENCHMARKS YET
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.
37 PAPERS • NO BENCHMARKS YET
For goal-oriented document-grounded dialogs, it often involves complex contexts for identifying the most relevant information, which requires better understanding of the inter-relations between conversations and documents. Meanwhile, many online user-oriented documents use both semi-structured and unstructured contents for guiding users to access information of different contexts. Thus, we create a new goal-oriented document-grounded dialogue dataset that captures more diverse scenarios derived from various document contents from multiple domains such ssa.gov and studentaid.gov. For data collection, we propose a novel pipeline approach for dialogue data construction, which has been adapted and evaluated for several domains.
34 PAPERS • NO BENCHMARKS YET
BLURB is a collection of resources for biomedical natural language processing. In general domains such as newswire and the Web, comprehensive benchmarks and leaderboards such as GLUE have greatly accelerated progress in open-domain NLP. In biomedicine, however, such resources are ostensibly scarce. In the past, there have been a plethora of shared tasks in biomedical NLP, such as BioCreative, BioNLP Shared Tasks, SemEval, and BioASQ, to name just a few. These efforts have played a significant role in fueling interest and progress by the research community, but they typically focus on individual tasks. The advent of neural language models such as BERTs provides a unifying foundation to leverage transfer learning from unlabeled text to support a wide range of NLP applications. To accelerate progress in biomedical pretraining strategies and task-specific methods, it is thus imperative to create a broad-coverage benchmark encompassing diverse biomedical tasks.
32 PAPERS • 2 BENCHMARKS
Worldtree is a corpus of explanation graphs, explanatory role ratings, and associated tablestore. It contains explanation graphs for 1,680 questions, and 4,950 tablestore rows across 62 semi-structured tables are provided. This data is intended to be paired with the AI2 Mercury Licensed questions.
32 PAPERS • NO BENCHMARKS YET
SCROLLS (Standardized CompaRison Over Long Language Sequences) is an NLP benchmark consisting of a suite of tasks that require reasoning over long texts. SCROLLS contains summarization, question answering, and natural language inference tasks, covering multiple domains, including literature, science, business, and entertainment. The dataset is made available in a unified text-to-text format and host a live leaderboard to facilitate research on model architecture and pretraining methods.
31 PAPERS • 1 BENCHMARK
DVQA is a synthetic question-answering dataset on images of bar-charts.
30 PAPERS • 1 BENCHMARK
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.
30 PAPERS • 2 BENCHMARKS
PlotQA is a VQA dataset with 28.9 million question-answer pairs grounded over 224,377 plots on data from real-world sources and questions based on crowd-sourced question templates. Existing synthetic datasets (FigureQA, DVQA) for reasoning over plots do not contain variability in data labels, real-valued data, or complex reasoning questions. Consequently, proposed models for these datasets do not fully address the challenge of reasoning over plots. In particular, they assume that the answer comes either from a small fixed size vocabulary or from a bounding box within the image. However, in practice this is an unrealistic assumption because many questions require reasoning and thus have real valued answers which appear neither in a small fixed size vocabulary nor in the image. In this work, we aim to bridge this gap between existing datasets and real world plots by introducing PlotQA. Further, 80.76% of the out-of-vocabulary (OOV) questions in PlotQA have answers that are not in a fixed
27 PAPERS • 5 BENCHMARKS
Natural Language Decathlon Benchmark (decaNLP) is a challenge that spans ten tasks: question answering, machine translation, summarization, natural language inference, sentiment analysis, semantic role labeling, zero-shot relation extraction, goal-oriented dialogue, semantic parsing, and commonsense pronoun resolution. The tasks as cast as question answering over a context.
27 PAPERS • NO BENCHMARKS YET
MathVista is a consolidated Mathematical reasoning benchmark within Visual contexts. It consists of three newly created datasets, IQTest, FunctionQA, and PaperQA, which address the missing visual domains and are tailored to evaluate logical reasoning on puzzle test figures, algebraic reasoning over functional plots, and scientific reasoning with academic paper figures, respectively. It also incorporates 9 MathQA datasets and 19 VQA datasets from the literature, which significantly enrich the diversity and complexity of visual perception and mathematical reasoning challenges within our benchmark. In total, MathVista includes 6,141 examples collected from 31 different datasets.
26 PAPERS • NO BENCHMARKS YET
WikiReading is a large-scale natural language understanding task and publicly-available dataset with 18 million instances. The task is to predict textual values from the structured knowledge base Wikidata by reading the text of the corresponding Wikipedia articles. The task contains a rich variety of challenging classification and extraction sub-tasks, making it well-suited for end-to-end models such as deep neural networks (DNNs).
We have created three new Reading Comprehension datasets constructed using an adversarial model-in-the-loop.
24 PAPERS • 2 BENCHMARKS
CaseHOLD (Case Holdings On Legal Decisions) is a law dataset comprised of over 53,000+ multiple choice questions to identify the relevant holding of a cited case. This dataset presents a fundamental task to lawyers and is both legally meaningful and difficult from an NLP perspective (F1 of 0.4 with a BiLSTM baseline). The citing context from the judicial decision serves as the prompt for the question. The answer choices are holding statements derived from citations following text in a legal decision. There are five answer choices for each citing text. The correct answer is the holding statement that corresponds to the citing text. The four incorrect answers are other holding statements.
23 PAPERS • 2 BENCHMARKS
The large-scale MUSIC-AVQA dataset of musical performance contains 45,867 question-answer pairs, distributed in 9,288 videos for over 150 hours. All QA pairs types are divided into 3 modal scenarios, which contain 9 question types and 33 question templates. Finally, as an open-ended problem of our AVQA tasks, all 42 kinds of answers constitute a set for selection.
22 PAPERS • 1 BENCHMARK
MultiDoc2Dial is a new task and dataset on modeling goal-oriented dialogues grounded in multiple documents. Most previous works treat document-grounded dialogue modeling as a machine reading comprehension task based on a single given document or passage. We aim to address more realistic scenarios where a goal-oriented information-seeking conversation involves multiple topics, and hence is grounded on different documents.
20 PAPERS • NO BENCHMARKS YET
The Cloze Test by Teachers (CLOTH) benchmark is a collection of nearly 100,000 4-way multiple-choice cloze-style questions from middle- and high school-level English language exams, where the answer fills a blank in a given text. Each question is labeled with a type of deep reasoning it involves, where the four possible types are grammar, short-term reasoning, matching/paraphrasing, and long-term reasoning, i.e., reasoning over multiple sentences
18 PAPERS • NO BENCHMARKS YET
CliCR is a new dataset for domain specific reading comprehension used to construct around 100,000 cloze queries from clinical case reports.
18 PAPERS • 1 BENCHMARK
A dataset on asking Questions for Lack of Clarity in open-domain information-seeking conversations. Qulac presents the first dataset and offline evaluation framework for studying clarifying questions in open-domain information-seeking conversational search systems.
Belebele is a multiple-choice machine reading comprehension (MRC) dataset spanning 122 language variants. This dataset enables the evaluation of mono- and multi-lingual models in high-, medium-, and low-resource languages. Each question has four multiple-choice answers and is linked to a short passage from the FLORES-200 dataset. The human annotation procedure was carefully curated to create questions that discriminate between different levels of generalizable language comprehension and is reinforced by extensive quality checks. While all questions directly relate to the passage, the English dataset on its own proves difficult enough to challenge state-of-the-art language models. Being fully parallel, this dataset enables direct comparison of model performance across all languages. Belebele opens up new avenues for evaluating and analyzing the multilingual abilities of language models and NLP systems.
17 PAPERS • NO BENCHMARKS YET
A new large-scale geometry problem-solving dataset - 3,002 multi-choice geometry problems - dense annotations in formal language for the diagrams and text - 27,213 annotated diagram logic forms (literals) - 6,293 annotated text logic forms (literals)
17 PAPERS • 1 BENCHMARK
TopiOCQA (pronounced Tapioca) is an open-domain conversational dataset with topic switches on Wikipedia. TopiOCQA contains 3,920 conversations with information-seeking questions and free-form answers. On average, a conversation in the dataset spans 13 question-answer turns and involves four topics (documents). TopiOCQA poses a challenging test-bed for models, where efficient retrieval is required on multiple turns of the same conversation, in conjunction with constructing valid responses using conversational history.
Video-and-Language Inference is the task of joint multimodal understanding of video and text. Given a video clip with aligned subtitles as premise, paired with a natural language hypothesis based on the video content, a model needs to infer whether the hypothesis is entailed or contradicted by the given video clip. The Violin dataset is a dataset for this task which consists of 95,322 video-hypothesis pairs from 15,887 video clips, spanning over 582 hours of video. These video clips contain rich content with diverse temporal dynamics, event shifts, and people interactions, collected from two sources: (i) popular TV shows, and (ii) movie clips from YouTube channels.
CLEVR-Ref+ is a synthetic diagnostic dataset for referring expression comprehension. The precise locations and attributes of the objects are readily available, and the referring expressions are automatically associated with functional programs. The synthetic nature allows control over dataset bias (through sampling strategy), and the modular programs enable intermediate reasoning ground truth without human annotators.
16 PAPERS • 2 BENCHMARKS
KaggleDBQA is a challenging cross-domain and complex evaluation dataset of real Web databases, with domain-specific data types, original formatting, and unrestricted questions.
16 PAPERS • 1 BENCHMARK
SUTD-TrafficQA (Singapore University of Technology and Design - Traffic Question Answering) is a dataset which takes the form of video QA based on 10,080 in-the-wild videos and annotated 62,535 QA pairs, for benchmarking the cognitive capability of causal inference and event understanding models in complex traffic scenarios. Specifically, the dataset proposes 6 challenging reasoning tasks corresponding to various traffic scenarios, so as to evaluate the reasoning capability over different kinds of complex yet practical traffic events.
Torque is an English reading comprehension benchmark built on 3.2k news snippets with 21k human-generated questions querying temporal relationships.
ConvQuestions is the first realistic benchmark for conversational question answering over knowledge graphs. It contains 11,200 conversations which can be evaluated over Wikidata. They are compiled from the inputs of 70 Master crowdworkers on Amazon Mechanical Turk, with conversations from five domains: Books, Movies, Soccer, Music, and TV Series. The questions feature a variety of complex question phenomena like comparisons, aggregations, compositionality, and temporal reasoning. Answers are grounded in Wikidata entities to enable fair comparison across diverse methods. The data gathering setup was kept as natural as possible, with the annotators selecting entities of their choice from each of the five domains, and formulating the entire conversation in one session. All questions in a conversation are from the same Turker, who also provided gold answers to the questions. For suitability to knowledge graphs, questions were constrained to be objective or factoid in nature, but no other r
15 PAPERS • NO BENCHMARKS YET
TV show Caption is a large-scale multimodal captioning dataset, containing 261,490 caption descriptions paired with 108,965 short video moments. TVC is unique as its captions may also describe dialogues/subtitles while the captions in the other datasets are only describing the visual content.
15 PAPERS • 1 BENCHMARK
FreebaseQA is a data set for open-domain QA over the Freebase knowledge graph. The question-answer pairs in this data set are collected from various sources, including the TriviaQA data set and other trivia websites (QuizBalls, QuizZone, KnowQuiz), and are matched against Freebase to generate relevant subject-predicate-object triples that were further verified by human annotators. As all questions in FreebaseQA are composed independently for human contestants in various trivia-like competitions, this data set shows richer linguistic variation and complexity than existing QA data sets, making it a good test-bed for emerging KB-QA systems.
14 PAPERS • NO BENCHMARKS YET
license: apache-2.0 tags: human-feedback size_categories: 100K<n<1M pretty_name: OpenAssistant Conversations
SQA3D is a dataset for embodied scene understanding, where an agent needs to comprehend the scene it situates from an first person's perspective and answer questions. The questions are designed to be situated, embodied and knowledge-intensive. We offer three different modalities to represent a 3D scene: 3D scan, egocentric video and BEV picture.
14 PAPERS • 2 BENCHMARKS
Who-did-What collects its corpus from news and provides options for questions similar to CBT. Each question is formed from two independent articles: an article is treated as context to be read and a separate article on the same event is used to form the query.
13 PAPERS • NO BENCHMARKS YET
GooAQ is a large-scale dataset with a variety of answer types. This dataset contains over 5 million questions and 3 million answers collected from Google. GooAQ questions are collected semi-automatically from the Google search engine using its autocomplete feature. This results in naturalistic questions of practical interest that are nonetheless short and expressed using simple language. GooAQ answers are mined from Google's responses to the collected questions, specifically from the answer boxes in the search results. This yields a rich space of answer types, containing both textual answers (short and long) as well as more structured ones such as collections.
11 PAPERS • NO BENCHMARKS YET
QAMPARI is an ODQA benchmark, where question answers are lists of entities, spread across many paragraphs. It was created by (a) generating questions with multiple answers from Wikipedia's knowledge graph and tables, (b) automatically pairing answers with supporting evidence in Wikipedia paragraphs, and (c) manually paraphrasing questions and validating each answer.
ConditionalQA is a Question Answering (QA) dataset that contains complex questions with conditional answers, i.e. the answers are only applicable when certain conditions apply.
10 PAPERS • 1 BENCHMARK
FM-IQA is a question-answering dataset containing over 150,000 images and 310,000 freestyle Chinese question-answer pairs and their English translations.
10 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.
Retrieval Question-Answering (ReQA) benchmark tests a model’s ability to retrieve relevant answers efficiently from a large set of documents.
We propose the first question-answering dataset driven by STEM theorems. We annotated 800 QA pairs covering 350+ theorems spanning across Math, EE&CS, Physics and Finance. The dataset is collected by human experts with very high quality. We provide the dataset as a new benchmark to test the limit of large language models to apply theorems to solve challenging university-level questions. We provide a pipeline in the following to prompt LLMs and evaluate their outputs with WolframAlpha.
BeaverTails is a dataset aimed at fostering research on safety alignment in large language models (LLMs). This dataset uniquely separates annotations of helpfulness and harmlessness for question-answering pairs, thus offering distinct perspectives on these crucial attributes. In total, the authors have compiled safety meta-labels for 30,207 question-answer (QA) pairs and gathered 30,144 pairs of expert comparison data for both
9 PAPERS • NO BENCHMARKS YET
CQASUMM is a dataset for CQA (Community Question Answering) summarization, constructed from the 4.4 million Yahoo! Answers L6 dataset. The dataset contains ~300k annotated samples.
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.