Loading...
Loading...
Text classification, translation, QA, summarization, dialogue, sentiment analysis, language modeling, text corpora
44,939 datasets
A set of 50,000 highly polar movie reviews for binary sentiment classification, with 25,000 reviews designated for training and 25,000 for testing. The dataset, created by Drane111, also includes additional unlabeled data for use. It was last updated on the platform in 2026.
Research data for the paper "Determination of Structures and Binding Energies of Europium Complexes Bound to Biologically Relevant Anions" by Felix Plasser, uploaded to figshare in April 2026. The 188.8 MB archive contains computational chemistry files organized into three main sections. It includes coordinate, input, and output files from optimization jobs performed using ORCA, QChem, and xTB software for various europium complexes.
CFB Gagetown military base boundaries were generated from data provided by Public Works and Government Services Canada (PWGSC). The dataset includes columns for area, names in English and French, geometry, and source identifiers. It was last updated on May 1, 2026, via the socrata platform.
Resampled data from the Low Energy Charged Particle experiment on Voyager 1 during its Jupiter encounter. The dataset consists of calibrated, sectored measurements of >26 keV electrons and >30 keV ions averaged to 15-minute intervals. It was produced by NASA and archived with the Planetary Data System.
The Geological Survey of Victoria identifies areas of known future interest to the extractive minerals industry. This dataset, updated in October 2023 and May 2024, is based on suitable geology and considers local planning schemes to guide policy development. It is provided by the Department of Energy, Environment and Climate Action.
Spatial points represent Department of Transport assets and coastal infrastructure in Western Australia, including water distribution, drainage, sewerage, electrical services, overbeach launches, boat mooring piles, signs, and dredged channels. Locations were generated from interpretation of aerial imagery by Geographic Information Management. The dataset is licensed under CC-BY and was last updated in March 2026.
Version 2.0 data from November 2009 onward, collected after the QuikSCAT antenna motor failure, provides averaged Level 1B Sigma-0 measurements and wind retrievals from repointed beams. The dataset is processed using the updated QNS2016a geophysical model function, incorporating ECMWF nowcast winds for calibration and applying a -0.25 dB Sigma-0 bias correction relative to Version 1. It supports cross-calibration of other scatterometers and extension of the Ku-band geophysical model function.
Revenue data for the Absentee Owner Surcharge (AOS), an additional land tax levied in Victoria, Australia. The State Revenue Office compiled this information starting from 1 July 2015, and it includes the number of transactions and associated revenue amounts. The dataset was last updated on 27 April 2026.
Voyager 2 spacecraft data from the Low Energy Charged Particle experiment during its Jupiter encounter. The dataset contains resampled, calibrated intensities of electrons (>13 keV) and ions (>24 keV) across eight angular sectors, averaged to 15-minute intervals. NASA produced this Version 1.1 dataset, which consolidates sector data into single records per channel.
A dataset from a 2026 figshare upload by Xue-ping Yang investigating the protective effects of bergamottin against cisplatin-induced acute kidney injury in mice and human renal tubular HK-2 cells. The data likely contains experimental results on cytotoxicity, renal function, inflammatory cytokines, lipid peroxidation, and BACE-1 expression.
Vietnamese legal documents and precedents collected from the anle.toaan.gov.vn portal of the Supreme People's Court. The corpus provides multi-level structured text with standardized Vietnamese spelling and full hierarchical markup from document to sentence level. The dataset ships in four configurations corresponding to different pipeline stages: documents, sentences, embeddings, and reduced representations.
Global sunlit Earth imagery collected by NASA's Terra spacecraft since its launch on December 18, 1999. The Multi-angle Imaging SpectroRadiometer (MISR) provides radiometrically and geometrically calibrated images in four spectral bands from nine different viewing angles. This data is designed to improve understanding of the Earth's environment and climate.
Myanmar government articles published by the Ministry of Information (MOI), systematically split into individual lines or sentences. The dataset is a derivative of the full-length article corpus, created by DatarrX and last updated on Hugging Face in May 2026.
Hua Song created a text file containing pre-generated equivalence sets for taxonomy entries, intended for use with large language models. The dataset is 7.7 KB in size and was last updated on May 14, 2026. It is shared under a CC-BY-4.0 license on the figshare platform.
A 320-row dataset created by author ayjays132, last updated on 2026-05-21. It is a small, curated text-to-image training dataset combining the original PhillMagine corpus with Gemini-synthesized samples. The dataset was used to train a universal text-conditioning adapter for diffusion models.
Gene expression data supporting the identification of NAD+ metabolism-related genes as biomarkers for pulmonary hypertension. The dataset originates from a 2026 bioinformatics study by Yanan Chu, published on figshare. Its small size of 8.4 KB indicates a focused, supplementary dataset.
Local Code Arena Telemetry contains raw evaluation metrics, execution logs, and syntax outputs from running the Mostly Basic Python Problems (MBPP) benchmark against Meta's Code Llama 7B model. The dataset, hosted by ShahzebKhoso on Hugging Face, documents baseline performance for early-generation specialized code engines. It was last updated on 2026-05-24.
syvb published the 'Qwen3-8B NLA — FULL parquets' dataset on Hugging Face on 2026-06-19. It contains neural network activation vectors recomputed for the Qwen3-8B model. The data is split into three configurations: av_sft, ar_sft, and rl, each with a different prompt schema.
The TGB Osborn (Koonamore) Vegetation Reserve in Australia contains data on the abundance of the Cassia (Senna artemesioides) plant species. The record covers 26 sampling instances from 1940 to 2014, tracking the number of individuals and their life stage (adult or juvenile) within a 7150 m2 area. This is version 2.1 of the data release, managed by the Terrestrial Ecosystem Research Network (TERN) and updated to comply with FAIR and TERN vocabulary standards.
Data underpinning a 2026 study comparing homologous recombination repair (HRR) deficiency and PARP-1 inhibitor sensitivity in MCPH1-deficient versus BRCA2-deficient human cells. Isobel G. Chapman generated this data to support figures in the associated publication. The dataset includes quantitative measurements from cell-based assays.