Loading...
Loading...
Text classification, translation, QA, summarization, dialogue, sentiment analysis, language modeling, text corpora
44,940 datasets
More than two million pages from 11,000 books printed in the Netherlands between 1781 and 1800. The collection originates from the Royal Library and the university libraries of Amsterdam and Leiden. Metadata is available under a CC0 license, while full-texts carry a Public Domain Mark.
Records detail city-initiated grass cutting on private properties under Chapter 66 lot abatement ordinances. The dataset includes fields for case status, complaint dates, addresses, and geographic coordinates, published by data.nola.gov. Data was last updated in April 2026.
Hakan Uzunlar published a study on figshare in April 2026. The dataset contains event-related theta spectral coherence measurements from electroencephalographic recordings of 106 participants, including patients with Alzheimer's disease, Parkinson's disease, and mild cognitive impairment, as well as healthy elderly controls. Participants were presented with angry, happy, and neutral facial expressions 60 times each in a pseudo-random order.
675 map sheets constitute the first large-scale, nationwide topographic survey of the German Empire. The Bundesamt für Kartographie und Geodäsie produced this map series, known as the General Staff Card, between 1878 and 1909. Each sheet covers an area of approximately 35 km by 28 km and was designed in polyhedral projection.
675 map sheets constitute the first large-scale topographic series covering the territory of the German Empire, completed in 1909. The maps were produced in a polyhedral projection, with each sheet covering an area of approximately 35 km by 28 km. The Bundesamt für Kartographie und Geodäsie is the authoritative organization for this historical cartographic work.
Raw evaluation metrics, execution telemetry logs, and structural syntax outputs from running the Mostly Basic Python Problems (MBPP) benchmark against the Qwen 3.5 2B base model. The dataset documents a zero-shot functional programming synthesis test under standard local execution bounds without active reasoning tokens. It was created by ShahzebKhoso and last updated on May 24, 2026.
The Cryo-EM SPA Workflow Records dataset, from the High Energy Accelerator Research Organization (KEK) Structural Biology Research Center, archives essential outputs and metadata from cryogenic electron microscopy single particle analysis workflows. It focuses on data generated by the RELION and CryoSPARC software packages, storing processing parameters and representative outputs to enable pipeline reconstruction while minimizing storage. This dataset is intended to enhance research transparency and support the development of AI algorithms for cryo-EM.
Radar echograms and derived measurements of glacier surface elevation, bed elevation, and ice thickness for select glaciers in Alaska and Northwestern Canada. The data was acquired by the University of Alaska Fairbanks High-Frequency Radar Sounder and is sourced from NASA IceBridge L1B products. It is hosted on AWS and requires an Earthdata Login account for access.
Airborne geophysical survey data collected over Antarctica, including vertical acceleration, free air gravity anomalies, radar echo strength, ice thickness, and surface elevation measurements. The data were gathered by scientists working on the ICECAP project, funded by the NSF and NERC with support from NASA Operation IceBridge. The dataset is hosted on AWS Open Data and is licensed under CC-BY-4.0.
Fermi LAT's 14-year catalog (4FGL-DR4) lists high-energy gamma-ray sources detected from 50 MeV to 1 TeV. The NASA HEASARC updated this database table in February 2024 using data from the Fermi Science Support Center. It provides updated spectral parameters, energy distributions, and light curves for all sources.
Jae Hyun Lee's thesis data includes confocal microscopy images of Wolffia expressing an ATP sensor and tabulated results from GUS assays, FRET ratio measurements, and NADP(H) assays. The dataset compares Wolffia australiana with the model plant Arabidopsis thaliana. It was last updated on May 15, 2026.
67,772 content files spanning 2,000 years of Catholic intellectual tradition, from the Church Fathers to the 20th century. The corpus aggregates 16 collections including the Patrologia Latina and Aquinas's Opera Omnia, totaling 35.9 GB of NLP-ready text. CatholicCorpus compiled this open-access resource, last updated on HuggingFace in May 2026.
A replication package reproduces the empirical implementation reported in Cid (2026). It contains cleaned input series, model-consistent datasets, scripts, and outputs for the Structural Debt Monitoring Framework. The repository implements structural variable construction, parameter calibration, state classification, and debt-dynamics diagnostics.
Australian seabed bathymetry compilation products delivered by Geoscience Australia. The dataset shows the spatial extent of compilations generated from sources including survey data, lidar, and interpolation as of June 2019. Each polygon includes attributes on data sources, product details, and access methods.
Structural statistics for HADDOCK-derived models of DENV-2 (Dengue virus type 2) proteins in complex with their interactors. The dataset compares non-phosphorylated, interface phosphorylated, and completely phosphorylated protein states. It was authored by Prakhar Agrawal and last updated on May 12, 2026.
261.2 KB of R code and metadata supports a study on persistent degradation of understory forest regeneration. The workflow includes scripts for data processing, statistical analysis, and figure generation, published under a CC-BY-4.0 license by Facundo Alvarez in April 2026. Raw forest inventory and Lidar data are not included but are available upon request for academic reproduction.
A collection of language examples illustrating equity-by-design principles for vulnerable populations within policy documents. The dataset was authored by Kimsey Zajac and last updated on 2026-05-19. It is a small dataset, 5.5 KB in size, available in XLS format under a CC-BY-4.0 license.
A paper by Tam Hunt proposes a mathematical heuristic for calculating spatial boundaries and phenomenal capacity of conscious resonating structures within General Resonance Theory. The work builds on a panpsychist framework developed over the last decade and compares the approach to Integrated Information Theory. The dataset likely contains the paper's text and associated theoretical models.
Raw evaluation metrics and execution telemetry logs from running the Mostly Basic Python Problems (MBPP) benchmark against the DeepSeek Coder 6.7B model. This dataset, authored by ShahzebKhoso, provides a specific anchor point for evaluating generational alignment improvements in code-generation architectures. It was last updated on May 24, 2026.
March to October 2024 fieldwork captured and released moths in different lighting environments around the University of Exeter's Penryn campus. The dataset records the time to move, type of movement, and direction of movement for each release to gauge the impact of lighting variation. The work was carried out by Maisy Inston, Charlie Rayner, and Jon Bennie.