12,168 datasets listed, ordered by the archive's paper count. Page 12 of 254: 48 shown of 12,168.
The archive holds 12,214 dataset rows; 12,168 are listed. 6 are withheld from every listing and count here as vandalised before snapshot (6 with contact-centre spam in the title, 0 with a spam description on a row that has no homepage, no paper and no papers counted; none with more than 1 paper, 0 with a benchmark), listed in withheld.json; 1 listed row carries a vandalised description, withheld on its page. This gate never withholds a row with a homepage or a paper that resolves, and a clean description; the content rules below withhold a row whose name is spam whatever else it carries. The gate is a phrase list: these are the rows it caught, not a claim that the rest is clean. Before that gate, the site's content rules withhold 40 more rows (invite-code, gambling, travel-booking, contact-centre and similar spam in the name or on a row with nothing real behind it); they have no page and are listed in withheld.json.
ORL (Our Database of Faces)
The ORL Database of Faces contains 400 images from 40 distinct subjects.
129 papers · 1 benchmark
TabFact is a large-scale dataset which consists of 117,854 manually annotated statements with regard to 16,573 Wikipedia tables, their relations are classified as ENTAILED and REFUTED.
129 papers · 2 benchmarks
VOT2018 is a dataset for visual object tracking.
129 papers · 1 benchmark
Europarl (European Parliament Proceedings Parallel Corpus)
A corpus of parallel text in 21 European languages from the proceedings of the European Parliament.
128 papers · 1 benchmark
The Free Music Archive (FMA) is a large-scale dataset for evaluating several tasks in Music Information Retrieval.
128 papers · 2 benchmarks
LFPW (Labeled Face Parts in the Wild)
The Labeled Face Parts in-the-Wild (LFPW) consists of 1,432 faces from images downloaded from the web using simple text queries on sites such as google.com, flickr.com, and yahoo.com.
128 papers · 0 benchmarks
The Meta-Dataset benchmark is a large few-shot learning benchmark and consists of multiple datasets of different data distributions.
128 papers · 2 benchmarks
We observe that satellite imagery is a powerful source of information as it contains more structured and uniform data, compared to traditional images.
127 papers · 1 benchmark
LogiQA consists of 8,678 QA instances, covering multiple types of deductive reasoning.
127 papers · 1 benchmark
NELL-995 KG Completion Dataset
127 papers · 2 benchmarks
SMAP (Soil Moisture Active Passive)
Soil Moisture Active Passive (SMAP) dataset is a dataset of soil samples and telemetry information using the Mars rover by NASA.
127 papers · 2 benchmarks
WNUT 2017 (WNUT 2017 Emerging and Rare entity recognition)
This shared task focuses on identifying unusual, previously-unseen entities in the context of emerging discussions.
127 papers · 2 benchmarks
WikiHow is a dataset of more than 230,000 article and summary pairs extracted and constructed from an online knowledge base written by different human authors.
127 papers · 2 benchmarks
BLiMP (Benchmark of Linguistic Minimal Pairs)
BLiMP is a challenge set for evaluating what language models (LMs) know about major grammatical phenomena in English.
126 papers · 0 benchmarks
DFDC (Deepfake Detection Challenge)
The DFDC (Deepfake Detection Challenge) is a dataset for deepface detection consisting of more than 100,000 videos.
126 papers · 1 benchmark
FBMS (Freiburg-Berkeley Motion Segmentation)
The Freiburg-Berkeley Motion Segmentation Dataset (FBMS-59) is an extension of the BMS dataset with 33 additional video sequences.
126 papers · 1 benchmark
LSMDC (Large Scale Movie Description Challenge)
This dataset contains 118,081 short video clips extracted from 202 movies.
126 papers · 3 benchmarks
PASCAL VOC 2007 is a dataset for image recognition.
126 papers · 13 benchmarks
SUN3D contains a large-scale RGB-D video database, with 8 annotated sequences.
126 papers · 0 benchmarks
Aff-Wild is a large-scale in-the-wild dataset for valence-arousal estimation from videos with a variety of head poses, illumination conditions and occlusions.
125 papers · 0 benchmarks
FreiHAND is a 3D hand pose dataset which records different hand actions performed by 32 people.
125 papers · 1 benchmark
The LUNA challenges provide datasets for automatic nodule detection algorithms using the largest publicly available reference database of chest CT scans, the LIDC-IDRI data set.
125 papers · 2 benchmarks
Click to add a brief description of the dataset (Markdown and LaTeX enabled).
125 papers · 0 benchmarks
The Yelp2018 dataset is adopted from the 2018 edition of the yelp challenge.
125 papers · 2 benchmarks
FER+ (Face Expression Recognition Plus dataset)
The FER+ dataset is an extension of the original FER dataset, where the images have been re-labelled into one of 8 emotion types: neutral, happiness, surprise, sadness, anger, disgust, fear, and contempt.
124 papers · 3 benchmarks
The MSRA-TD500 dataset is a text detection dataset that contains 300 training images and 200 test images.
124 papers · 1 benchmark
The Microsoft Academic Graph is a heterogeneous graph containing scientific publication records, citation relationships between those publications, as well as authors, institutions, journals, conferences, and fields of study.
124 papers · 0 benchmarks
NCLT (North Campus Long-Term Vision and LiDAR)
The NCLT dataset is a large scale, long-term autonomy dataset for robotics research collected on the University of Michigan’s North Campus.
124 papers · 0 benchmarks
CARER (Contextualized Affect Representations for Emotion Recognition)
CARER is an emotion dataset collected through noisy labels, annotated via distant supervision as in (Go et al., 2009).
123 papers · 2 benchmarks
JFT-300M is an internal Google dataset used for training image classification models.
123 papers · 1 benchmark
PubLayNet is a dataset for document layout analysis by automatically matching the XML representations and the content of over 1 million PDF articles that are publicly available on PubMed Central.
123 papers · 1 benchmark
BBQ (Bias Benchmark for QA)
Bias Benchmark for QA (BBQ) is a dataset consisting of question-sets constructed by the authors that highlight attested social biases against people belonging to protected classes along nine different social dimensions relevant for U.S.…
122 papers · 0 benchmarks
LibriMix is an open-source alternative to wsj0-2mix.
122 papers · 1 benchmark
MAWPS (MAth Word ProblemS)
MAWPS is an online repository of Math Word Problems, to provide a unified testbed to evaluate different algorithms.
122 papers · 1 benchmark
Multi-News, consists of news articles and human-written summaries of these articles from the site newser.com.
122 papers · 5 benchmarks
ETHD is a multi-view stereo benchmark / 3D reconstruction benchmark that covers a variety of indoor and outdoor scenes.
121 papers · 3 benchmarks
The GENIA corpus is the primary collection of biomedical literature compiled and annotated within the scope of the GENIA project.
121 papers · 7 benchmarks
Permuted MNIST is an MNIST variant that consists of 70,000 images of handwritten digits from 0 to 9, where 60,000 images are used for training, and 10,000 images for test.
121 papers · 1 benchmark
The dataset for the SemEval-2010 Task 8 is a dataset for multi-way classification of mutually exclusive semantic relations between pairs of nominals.
121 papers · 1 benchmark
GTEA (Georgia Tech Egocentric Activity)
The Georgia Tech Egocentric Activities (GTEA) dataset contains seven types of daily activities such as making sandwich, tea, or coffee.
120 papers · 2 benchmarks
Raindrop is a set of image pairs, where each pair contains exactly the same background scene, yet one is degraded by raindrops and the other one is free from raindrops.
120 papers · 1 benchmark
SHAPES (Swarm Heuristics based Adaptive and Penalized Estimation of Splines)
SHAPES is a dataset of synthetic images designed to benchmark systems for understanding of spatial and logical relations among multiple objects.
120 papers · 1 benchmark
SIQA (Social Interaction QA)
Social Interaction QA (SIQA) is a question-answering benchmark for testing social commonsense intelligence.
120 papers · 1 benchmark
SimpleQuestions is a large-scale factoid question answering dataset.
120 papers · 2 benchmarks
The 'shape bias' dataset was introduced in Geirhos et al. (ICLR 2019) and consists of 224x224 images with conflicting texture and shape information (e.g., cat shape with elephant texture).
120 papers · 1 benchmark
SEED (SJTU Emotion EEG Dataset)
The SEED dataset contains subjects' EEG signals when they were watching films clips.
119 papers · 4 benchmarks
The MAESTRO dataset contains over 200 hours of paired audio and MIDI recordings from ten years of International Piano-e-Competition.
118 papers · 1 benchmark
VATEX is multilingual, large, linguistically complex, and diverse dataset in terms of both video and natural language descriptions.
118 papers · 3 benchmarks
Paper counts and descriptions are the archive's, frozen 2025-07-28; no citation counts, no stars, no trending. Sorting by "most cited" or "newest" was a live-site feature the archive does not carry.