Loading...
Loading...
Text classification, translation, QA, summarization, dialogue, sentiment analysis, language modeling, text corpora
46,421 datasets
2,502 unique, synthetic Python coding examples designed to train large language models for expert-level programming. The dataset, created by author '11-47', focuses on advanced topics like metaprogramming and asyncio. It was last updated on May 17, -2026.
May 2026 saw the release of a master lexical dataset for the Sindhi language, containing 223,342 entries drawn from multiple dictionary and terminology sources. It was prepared and curated by Amar Fayaz Buriro for SindhiLanguage.org. The dataset is structured and normalized into CSV, JSONL, and SQLite formats for developer and AI use.
TeichAI's dataset contains raw agent trace files generated using the teich tool. The dataset includes 4,006 JSONL files where all assistant responses were generated by the deepseek/deepseek-v4-pro model. It was last updated on May 12, 2026.
A mixed-methods study by Sana Siddiqui, last updated in April 2026, explores barriers to Employee Assistance Program (EAP) usage. It includes 30 semi-structured interviews with EAP-eligible employees aged 18ā65 in Canada and the United States who had not used EAP services. The data was analyzed thematically and used to train a BERT-based NLP model to classify quotes by country.
Lobbying registrations filed with the Office of the Registrar of Lobbyists and its successor, the Office of the Commissioner of Lobbying, from January 31, 1996 onward. The dataset includes a primary file and secondary CSV files to represent one-to-many relationships for certain registration elements, accompanied by a data dictionary for definitions.
Canadian federal data discloses oral and arranged communications between lobbyists and Designated Public Office Holders. Reports are filed monthly with the Office of the Commissioner of Lobbying under the Lobbying Act, which came into force on July 2, 2008. The dataset includes primary and secondary CSV files to handle one-to-many relationships for some data elements.
A full factorial parameter study on 5G NR conducted in the University of Wuerzburg 5G campus network. The dataset contains aggregated measurements of channel quality, energy consumption, and traffic KPIs for two gNB types and two modems across varied configurations. Lukas Kilian Schumann published this dataset on April 23, 2026.
100 million high-quality, diverse images comprise the Giant Permissive Image Corpus (GPIC), a dataset for visual generation. The dataset was introduced by researchers from Stanford University, with the associated paper available on arXiv. The dataset page was last updated on 2026-05 06.
HISCALE Low-Energy Magnetic Spectrometer ion counts at 120 degrees from the Ulysses spacecraft spin axis. The data covers the Jupiter encounter from 1991-12-31 to 1992-02-16, including 1-hour averaged inbound cruise data and 15-minute averaged encounter data. The dataset was produced by the National Aeronautics and Space Administration.
Multibeam sonar bathymetry data from the South-west Corner and Perth Canyon Marine Parks, acquired in March-April 2022. The dataset contains five GeoTIFF files at 64m and 128m resolutions, processed from Kongsberg EM304 data. The survey was led by Dr. Todd Bond of the Minderoo-UWA Deep-Sea Research Centre aboard the MV Pangaea Ocean Explorer.
Supplementary Data 1-4 from a study authenticating the Cambrian origin of Bryozoa. The collection includes micrometer-scale measurements of fossil zooids, a character taxon matrix in NEXUS format, and a dataset of 50 morphological characters across 21 taxa for phylogenetic analysis. Authored by Baopeng Song and published on figshare in April 2026 under a CC-BY-4.0 license.
Australia's marine jurisdiction seabed geomorphic features are described by datasets from Geoscience Australia. The jurisdiction covers the Exclusive Economic Zone, including offshore islands and territories. Features were identified using the best available bathymetric data and mapped at a scale of 1:5,000,000.
Australia's national-scale mining sector is analyzed through a review of 33 spatial ESG datasets across six themes: people, land uses, water resources, extreme events, nature conservation, and governance. The dataset, associated with a 2025 research paper, was published by Geoscience Australia Data. It provides a foundation for ESG mapping to inform decision-making for the energy transition.
Bikash Ranjan Nayak's research dataset analyzes mineral chemistry and trace elements from the Spongtang ophiolite in Ladakh. It details elemental pathways and paleo-redox conditions during serpentinization across sub-greenschist to eclogite facies. The 20.6 MB document was last updated in April 2026.
A dataset for supervised fine-tuning of language models, generated by the ML Intern agent on the Hugging Face Hub. The repository was created by the user dignity045 and was last updated on June 4, 2026. The specific content, size, and structure of the data are not detailed in the available metadata.
2017 data from the Maryland Department of Transportation State Highway Administration details the linear geometric features of state highways classified as Primary or Secondary systems. This dataset is used for general planning and funding purposes by various government agencies and is updated annually. It represents the inventory direction side of roadways and is not a complete representation of all geometry.
A review project by Godwin Chakkappan, last updated on 2026-05-27, aims to interpret and identify gaps in knowledge regarding data security and Explainable AI technologies used for decision-making in the finance industry. It examines methodologies, technologies, and approaches for optimal decision-making alongside data security.
54.5 KB of code implements a system for automated English writing scoring and feedback using deep semantic analysis. The system, created by Shuling Yang and uploaded in April 2026, processes the ASAP-AES dataset to predict overall essay scores and four dimensional scores. It includes modules for error detection and personalized feedback generation.
Reporting of publicly funded top incomes in accordance with the Public Disclosure-Funded Top Income Act (Wopt) and income data and the reporting of directors of healthcare institutions from the 2011 annual reports. The report was presented to the States General on 8 January 2013. The dataset likely contains tabular records of income disclosures from institutions financed from public funds.
Geoscience Australia's web service maps geomorphic features across Australia's marine jurisdiction, including its Exclusive Economic Zone and offshore territories. Features are identified using the best available bathymetric data and are generally mapped at a scale of 1:5,000,000. The service is provided by the Australian Ocean Data Network.