Loading...
Loading...
News corpora, social media analysis, movie/music metadata, sports data, cultural datasets, misinformation
11,230 datasets
A 2012 dataset of full-service grocery store locations in the District of Columbia, compiled by the Office of the Chief Technology Officer. It uses the Alcoholic Beverage and Cannabis Administration's licensing criteria and was cross-referenced with a 2018 Food System Assessment. Data is confirmed through government datasets, commercial sources, and site visits.
An ATOM feed provides access to the development plan for sports facilities in the municipality of Schwanewede, structured according to the INSPIRE PLU (Planning Land Use) directive. The dataset is provided by the Bundesamt für Kartographie und Geodäsie (Federal Agency for Cartography and Geodesy) and was last updated on August 9, 2025.
A WFS service of the urban development plan ‘Alter Sportplatz’ for the municipality of Meckesheim, transformed according to INSPIRE standards. The data is provided by the Bundesamt für Kartographie und Geodäsie and was last updated on August 7, 2025.
Deep Noise Suppression Challenge datasets contain clean speech and noise clips for audio enhancement research. The repository includes inference scripts and an ONNX model for a baseline Noise Suppression Network. The dataset was uploaded by ltnghia on July 28, 2025.
FAVOR-Bench is a new benchmark for fine-grained video motion understanding released in March 2025. It spans both ego-centric and third-person video perspectives and includes both close-ended QA and open-ended descriptive evaluation tasks. The dataset was created by author zl2048 and last updated on Hugging Face in August 2025.
552,821 unique text samples with 570,457 emotion label assignments form this large-scale, multilabel dataset. It aggregates data from six prominent emotion datasets: MELD, GoEmotions, TwitterEmotion, ISEAR, SemEval, and Crowdflower. The dataset was created by 'cirimus' and was last updated on July 28, 2025.
Egomask is a benchmark dataset for mask annotation in egocentric vision. The dataset's structure includes separate directories for training data and benchmark evaluation subsets. It was uploaded by XuuuXYZ and last updated on August 4, 2025.
A dataset of synthetically-generated question-answer items created from filtered Reddit data. The dataset was created by AllenAI and was last updated on July 28, 2025. Its creation was motivated by findings from the Dolma project regarding the limited benefit of thread-level context in Reddit data for certain benchmarks.
Open-world-agents provides the VPT Owamcap dataset, a conversion of the Video PreTraining (VPT) Minecraft dataset into the OWAMcap format. It contains 26,322 samples with a total volume of 5.2 terabytes, each pairing a video file with corresponding action logs. The dataset was last updated on July 10, 2025.
55,521 high-resolution images across 15 sequences depict a fixed-wing aircraft on a collision course with a stationary camera. The dataset, created by Martin, Riseley, and Ford in 2022, includes ground truth labels and video sequences. It is hosted on Hugging Face and was last updated on July 14, 2025.
Peterborough City Council's land and property assets are detailed in this dataset. The information is correct as of the upload date, with annual updates typically provided around mid-April. The data is provided by the Government Digital Service via the eu_open_data platform.
Approximately 114,000 user reviews collected from over 150 movies on IMDb. The dataset is organized by movie, with each movie's reviews stored in a separate JSON file identified by its IMDb ID. Author Daksh0505 uploaded the dataset to Hugging Face, with a last recorded update on 2025-07-16.
A dataset named 'Expresso Read With Rebuilt Caption' was published on the Hugging Face platform by author 'morateng'. It was last updated on September 9, 2025. The platform tags suggest it contains text data and is stored in Parquet format.
maafiah's Idiomsresearch repository hosts data and links for research on idiomatic expressions. The repository was last updated on 2025-09-18. Its specific contents and scale are not detailed in the available metadata.
SVDQuant datasets are used for benchmarking the SVDQuant method, which absorbs outliers by low-rank components for 4-bit diffusion models. The dataset is hosted by MIT-HAN-Lab and was last updated on July 21, 2025. It is associated with a research paper and code repository for the SVDQuant technique.
2,273 professional football players are profiled with performance statistics from the 2024-2025 season across top European leagues. Sourced from FBref by 3zden, the data includes both traditional metrics and advanced analytics like xG and progressive actions.
CIDAR is a culturally relevant instruction dataset for Arabic containing 10,000 instruction-output pairs. The dataset was created by the arbml organization, with the latest update recorded on July 1, —2025.
Fastiv, Ukraine, is the geographic scope for this dataset on issued building passports. The data was published by the States site of Ukraine on the eu_open_data platform and was last updated on 2025-08-05. The specific content and scale of the records are not detailed in the available metadata.
A collection of over 100,000 synthetically generated passport images created by ud-biometrics and updated in July 2025. The dataset includes document layouts and photos from more than 100 countries. It is designed for training neural networks in identity verification and fraud detection tasks.
The dataset contains a list of construction passports for land plot development issued by the Department of Architecture, Urban Planning and Construction of the Korsun-Shevchenkivskyi City Council. It likely includes information about customers, the authorized body, issued documents, and construction projects. The data was last updated on July 9, 2025.