The Implicit Hate corpus is a dataset for hate speech detection with fine-grained labels for each message and its implication. This dataset contains 22,056 tweets from the most prominent extremist groups in the United States; 6,346 of these tweets contain implicit hate speech.
43 PAPERS • NO BENCHMARKS YET
MusicCaps is a dataset composed of 5.5k music-text pairs, with rich text descriptions provided by human experts. For each 10-second music clip, MusicCaps provides:
43 PAPERS • 1 BENCHMARK
TurkCorpus, a dataset with 2,359 original sentences from English Wikipedia, each with 8 manual reference simplifications. The dataset is divided into two subsets: 2,000 sentences for validation and 359 for testing of sentence simplification models.
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
EmoContext consists of three-turn English Tweets. The emotion labels include happiness, sadness, anger and other.
EmotionLines contains a total of 29245 labeled utterances from 2000 dialogues. Each utterance in dialogues is labeled with one of seven emotions, six Ekman’s basic emotions plus the neutral emotion. Each labeling was accomplished by 5 workers, and for each utterance in a label, the emotion category with the highest votes was set as the label of the utterance. Those utterances voted as more than two different emotions were put into the non-neutral category. Therefore the dataset has a total of 8 types of emotion labels, anger, disgust, fear, happiness, sadness, surprise, neutral, and non-neutral.
English Web Treebank is a dataset containing 254,830 word-level tokens and 16,624 sentence-level tokens of webtext in 1174 files annotated for sentence- and word-level tokenization, part-of-speech, and syntactic structure. The data is roughly evenly divided across five genres: weblogs, newsgroups, email, reviews, and question-answers. The files were manually annotated following the sentence-level tokenization guidelines for web text and the word-level tokenization guidelines developed for English treebanks in the DARPA GALE project. Only text from the subject line and message body of posts, articles, messages and question-answers were collected and annotated.
42 PAPERS • NO BENCHMARKS YET
The Visual Spatial Reasoning (VSR) corpus is a collection of caption-image pairs with true/false labels. Each caption describes the spatial relation of two individual objects in the image, and a vision-language model (VLM) needs to judge whether the caption is correctly describing the image (True) or not (False).
COCO-O(ut-of-distribution) contains 6 domains (sketch, cartoon, painting, weather, handmake, tattoo) of COCO objects which are hard to be detected by most existing detectors. The dataset has a total of 6,782 images and 26,624 labelled bounding boxes.
41 PAPERS • 1 BENCHMARK
The RWC (Real World Computing) Music Database is a copyright-cleared music database (DB) that is available to researchers as a common foundation for research. It contains around 100 complete songs with manually labeled section boundaries. For the 50 instruments, individual sounds at half-tone intervals were captured with several variations of playing styles, dynamics, instrument manufacturers and musicians.
41 PAPERS • NO BENCHMARKS YET
The SCUT-CTW1500 dataset contains 1,500 images: 1,000 for training and 500 for testing. In particular, it provides 10,751 cropped text instance images, including 3,530 with curved text. The images are manually harvested from the Internet, image libraries such as Google Open-Image, or phone cameras. The dataset contains a lot of horizontal and multi-oriented text.
41 PAPERS • 3 BENCHMARKS
We introduce a dataset of 147 object categories containing over 6000 images that are suitable for the few-shot counting task. We collected and annotated images ourselves. Our dataset consists of 6135 images across a di- verse set of 147 object categories, from kitchen utensils and office stationery to vehicles and animals. The object count in our dataset varies widely, from 7 to 3731 objects, with an average count of 56 objects per image. In each image, each object instance is annotated with a dot at its approxi- mate center. In addition, three object instances are selected randomly as exemplar instances; these exemplars are also annotated with axis-aligned bounding boxes.
40 PAPERS • 3 BENCHMARKS
A large-scale MultiLingual SUMmarization dataset. Obtained from online newspapers, it contains 1.5M+ article/summary pairs in five different languages -- namely, French, German, Spanish, Russian, Turkish. Together with English newspapers from the popular CNN/Daily mail dataset, the collected data form a large scale multilingual dataset which can enable new research directions for the text summarization community.
40 PAPERS • 5 BENCHMARKS
Dataset Summary
40 PAPERS • 1 BENCHMARK
UAV-Human is a large dataset for human behavior understanding with UAVs. It contains 67,428 multi-modal video sequences and 119 subjects for action recognition, 22,476 frames for pose estimation, 41,290 frames and 1,144 identities for person re-identification, and 22,263 frames for attribute recognition. The dataset was collected by a flying UAV in multiple urban and rural districts in both daytime and nighttime over three months, hence covering extensive diversities w.r.t subjects, backgrounds, illuminations, weathers, occlusions, camera motions, and UAV flying attitudes. This dataset can be used for UAV-based human behavior understanding, including action recognition, pose estimation, re-identification, and attribute recognition.
ECHR is an English legal judgment prediction dataset of cases from the European Court of Human Rights (ECHR). The dataset contains ~11.5k cases, including the raw text.
39 PAPERS • 1 BENCHMARK
RSTPReid contains 20505 images of 4,101 persons from 15 cameras. Each person has 5 corresponding images taken by different cameras with complex both indoor and outdoor scene transformations and backgrounds in various periods of time, which makes RSTPReid much more challenging and more adaptable to real scenarios. Each image is annotated with 2 textual descriptions. For data division, 3701 (index < 18505), 200 (18505 <= index < 19505) and 200 (index >= 19505) identities are utilized for training, validation and testing, respectively (Marked by item 'split' in the JSON file). Each sentence is no shorter than 23 words.
39 PAPERS • 2 BENCHMARKS
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
XD-Violence is a large-scale audio-visual dataset for violence detection in videos.
This dataset contains a large number of segmented nuclei images. The images were acquired under a variety of conditions and vary in the cell type, magnification, and imaging modality (brightfield vs. fluorescence). The dataset is designed to challenge an algorithm's ability to generalize across these variations.
38 PAPERS • 1 BENCHMARK
Assembly101 is a new procedural activity dataset featuring 4321 videos of people assembling and disassembling 101 "take-apart" toy vehicles. Participants work without fixed instructions, and the sequences feature rich and natural variations in action ordering, mistakes, and corrections. Assembly101 is the first multi-view action dataset, with simultaneous static (8) and egocentric (4) recordings. Sequences are annotated with more than 100K coarse and 1M fine-grained action segments, and 18M 3D hand poses. We benchmark on three action understanding tasks: recognition, anticipation and temporal segmentation. Additionally, we propose a novel task of detecting mistakes. The unique recording format and rich set of annotations allow us to investigate generalization to new toys, cross-view transfer, long-tailed distributions, and pose vs. appearance. We envision that Assembly101 will serve as a new challenge to investigate various activity understanding problems.
38 PAPERS • 4 BENCHMARKS
BEAT has i) 76 hours, high-quality, multi-modal data captured from 30 speakers talking with eight different emotions and in four different languages, ii) 32 millions frame-level emotion and semantic relevance annotations. Our statistical analysis on BEAT demonstrates the correlation of conversational gestures with \textit{facial expressions}, \textit{emotions}, and \textit{semantics}, in addition to the known correlation with \textit{audio}, \textit{text}, and \textit{speaker identity}. Based on this observation, we propose a baseline model, \textbf{Ca}scaded \textbf{M}otion \textbf{N}etwork \textbf{(CaMN)}, which consists of above six modalities modeled in a cascaded architecture for gesture synthesis. To evaluate the semantic relevancy, we introduce a metric, Semantic Relevance Gesture Recall (\textbf{SRGR}). Qualitative and quantitative experiments demonstrate metrics' validness, ground truth data quality, and baseline's state-of-the-art performance. To the best of our knowledge,
The BUCC mining task is a shared task on parallel sentence extraction from two monolingual corpora with a subset of them assumed to be parallel, and that has been available since 2016. For each language pair, the shared task provides a monolingual corpus for each language and a gold mapping list containing true translation pairs. These pairs are the ground truth. The task is to construct a list of translation pairs from the monolingual corpora. The constructed list is compared to the ground truth, and evaluated in terms of the F1 measure.
CodeContests is a competitive programming dataset for machine-learning. This dataset was used when training AlphaCode.
ESD is an Emotional Speech Database for voice conversion research. The ESD database consists of 350 parallel utterances spoken by 10 native English and 10 native Chinese speakers and covers 5 emotion categories (neutral, happy, angry, sad and surprise). More than 29 hours of speech data were recorded in a controlled acoustic environment. The database is suitable for multi-speaker and cross-lingual emotional voice conversion studies.
38 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.
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.
37 PAPERS • NO BENCHMARKS YET
We release E-commerce Dialogue Corpus, comprising a training data set, a development set and a test set for retrieval based chatbot. The statistics of E-commerical Conversation Corpus are shown in the following table.
37 PAPERS • 1 BENCHMARK
Legal General Language Understanding Evaluation (LexGLUE) benchmark is a collection of datasets for evaluating model performance across a diverse set of legal NLU tasks in a standardized way.
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.
Tatoeba is a free collection of example sentences with translations geared towards foreign language learners. It is available in more than 400 languages. Its name comes from the Japanese phrase “tatoeba” (例えば), meaning “for example”. It is written and maintained by a community of volunteers through a model of open collaboration. Individual contributors are known as Tatoebans.
37 PAPERS • 26 BENCHMARKS
WMT 2018 is a collection of datasets used in shared tasks of the Third Conference on Machine Translation. The conference builds on a series of twelve previous annual workshops and conferences on Statistical Machine Translation.
37 PAPERS • 6 BENCHMARKS
FEVEROUS (Fact Extraction and VERification Over Unstructured and Structured information) is a fact verification dataset which consists of 87,026 verified claims. Each claim is annotated with evidence in the form of sentences and/or cells from tables in Wikipedia, as well as a label indicating whether this evidence supports, refutes, or does not provide enough information to reach a verdict.
36 PAPERS • NO BENCHMARKS YET
Gait3D is a large-scale 3D representation-based gait recognition dataset. It contains 4,000 subjects and over 25,000 sequences extracted from 39 cameras in an unconstrained indoor scene.
36 PAPERS • 2 BENCHMARKS
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.
There exist previous works [6, 10] that constructed referring segmentation datasets for videos. Gavrilyuk et al. [6] extended the A2D [33] and J-HMDB [9] datasets with natural sentences; the datasets focus on describing the ‘actors’ and ‘actions’ appearing in videos, therefore the instance annotations are limited to only a few object categories corresponding to the dominant ‘actors’ performing a salient ‘action’. Khoreva et al. [10] built a dataset based on DAVIS [25], but the scales are barely sufficient to learn an end-to-end model from scratch
36 PAPERS • 3 BENCHMARKS
The SemEval-2018 hypernym discovery evaluation benchmark (Camacho-Collados et al. 2018) contains three domains (general, medical and music) and is also available in Italian and Spanish (not in this repository). For each domain a target corpus and vocabulary (i.e. hypernym search space) are provided. The dataset contains both concepts (e.g. dog) and entities (e.g. Manchester United) up to trigrams.
DialogRE is the first human-annotated dialogue-based relation extraction dataset, containing 1,788 dialogues originating from the complete transcripts of a famous American television situation comedy Friends. The are annotations for all occurrences of 36 possible relation types that exist between an argument pair in a dialogue. DialogRE is available in English and Chinese.
35 PAPERS • 1 BENCHMARK
EBM-NLP annotates PICO (Participants, Interventions, Comparisons and Outcomes) spans in clinical trial abstracts. The corresponding PICO Extraction task aims to identify the spans in clinical trial abstracts that describe the respective PICO elements.
The KIT Motion-Language is a dataset linking human motion and natural language.
35 PAPERS • 2 BENCHMARKS
node classification on genius
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.
34 PAPERS • 2 BENCHMARKS
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
The Open Entity dataset is a collection of about 6,000 sentences with fine-grained entity types annotations. The entity types are free-form noun phrases that describe appropriate types for the role the target entity plays in the sentence. Sentences were sampled from Gigaword, OntoNotes and web articles. On average each sentence has 5 labels.
PanLex translates words in thousands of languages. Its database is panlingual (emphasizes coverage of every language) and lexical (focuses on words, not sentences).
SciCite is a dataset of citation intents that addresses multiple scientific domains and is more than five times larger than ACL-ARC.
34 PAPERS • 3 BENCHMARKS
This dataset contains product reviews and metadata from Amazon, including 142.8 million reviews spanning May 1996 - July 2014.
33 PAPERS • 6 BENCHMARKS
Composed Image Retrieval (or, Image Retreival conditioned on Language Feedback) is a relatively new retrieval task, where an input query consists of an image and short textual description of how to modify the image.
33 PAPERS • 3 BENCHMARKS