Loading...
Loading...
Image classification, object detection, segmentation, face recognition, OCR, image generation, video understanding
17,570 datasets
MLLM-CITBench provides a benchmark for evaluating multimodal large language models across seven distinct task domains. The dataset includes independent training and test splits for tasks in OCR, art, financial policy, math, medicine, numerical reasoning, and science. It was created by yueluoshuangtian and last updated on Hugging Face in October 2025.
A dataset containing coordinate annotation data for text lines in Tibetan documents. It includes features related to text lines, image details, and processing methods used for data annotation, created by openpecha and last updated on 2025-10-13 09:41:33. The dataset is hosted on the Hugging Face platform.
70 interview records with Lebanese militants who died fighting for ISIS and other radical organizations in Syria and Iraq. The dataset was created by Hicham Bou Nassif and harvested by QDR Dataverse. It was last updated on October 20, 2025.
Research data on energy democracy initiatives and their transition narratives in northeastern North America. Matthew Burke authored this dataset, which was last updated on October 20, 2025. The data supports initiative-based practice and learning for renewable energy transitions.
Replication data from a computational study investigating artificial water channels. The dataset likely contains results from all-atom molecular dynamics simulations used to calculate water permeability for ligand-appended pillar[n]arene channels with varying ring-sizes and side-chain lengths. The data was authored by Tyler J. Duncan and last updated on October 15, 2025.
ExpVid is a benchmark dataset for evaluating Multimodal Large Language Models on scientific experiment videos. It comprises 390 lab experiment videos spanning 13 disciplines and includes 10 tasks across 3 levels of understanding. The dataset was created by OpenGVLab and was last updated on October 14, 2025.
LTDv2 is a single-scene thermal multi-object detection dataset created by vapaau. It contains over 1 million frames extracted from video clips spanning 8 months, with more than 6.8 million annotated bounding boxes for 4 object classes. The data encompasses several seasons, weather conditions, and day-night cycles.
Designed to evaluate the Optical Character Recognition capabilities of Multimodal Large Language Models in video scenarios. It was created by DogNeverSleep for the associated research paper and was last updated in October 2025.
A Netherlands-based survey of researchers working at universities, university medical centers, applied sciences universities, NWO or KNAW institutes, or public knowledge organizations. The data was collected in Q4 2017 for the Ministry of Education, Culture and Science, building upon a 2013 survey. The dataset is authored by J. de Jonge and curated by DANS Data Station Social Sciences and Humanities.
VGGFace2 is a large-scale dataset of facial images for recognition research. It contains over 3.1 million images of more than 9,000 subjects, created by the Visual Geometry Group at the University of Oxford. The dataset was released to advance face recognition model training.
ImageNet is a large-scale hierarchical image database organized according to the WordNet noun hierarchy. It contains over 14 million hand-annotated images across more than 20,000 categories. The dataset was created by researchers at Stanford and Princeton and is widely known for the annual ImageNet Large Scale Visual Recognition Challenge (ILSVRC) that ran from 2010 to 2017.
Bangla OCR is a dataset for optical character recognition in the Bengali language, published on the HuggingFace platform. The dataset was uploaded by author MdBayazid01 and was last updated on November 23, 2025. Its specific content, scale, and collection methodology are not detailed in the available metadata.
Structured Turkish chat data about public institutions, ministries, state bodies, official symbols, and historical figures in Turkey. The dataset was automatically extracted from Turkish Wikipedia, a reliable and neutral source, and converted into a format suitable for fine-tuning large language models (LLMs). It was created by author kaan39 and last updated on October 14, 2025.
Image and text data from Wikipedia Simple English, with crowdsourced annotations. The dataset size is categorized as between 100K and 1M instances, and it is stored in Parquet format.
27,000 synchronized frames of 16-layer LiDAR, RGB camera, and 4D radar data were collected across multiple buildings at the University of British Columbia. The Indoor FireRescue Rada (IFR) dataset includes raw ADC data, point clouds, and 3D bounding box annotations for objects like doors, chairs, and fire hydrants. This dataset was created by author yysd123 and last updated on Hugging Face in October 2025.
1,237 respondents with jobs and direct supervisors provided data on virtuous leadership, trust, and work-related well-being. The dataset was collected via the Prolific crowdsourcing platform in January 2019. M. Hendriks authored this dataset, which is hosted by the DANS Data Station Social Sciences and Humanities Collection.
From 1978 to 2004, this dataset provides a view of the consumption expenditures of private households in the Netherlands, detailing their size, composition, and financing. It is a protected microdata file from Statistics Netherlands (CBS), processed to prevent the identification of individuals or households. The data is hosted by the DANS Data Station Social Sciences and Humanities Collection.
A 1965 survey of classical studies students at the University of Amsterdam, conducted by the Scaliger study committee. The data likely contains information on study behavior, opinions on study programs, and organizational memberships. Background variables include basic characteristics, housing situation, occupation, income, capital assets, education, and organizational membership.
An oral history interview with a highly educated Sinti woman born in 1961, conducted in 2020. The interview covers her educational journey, including leaving school after primary education and later pursuing vocational training, as well as her significant role in a former self-organization. The data is part of the DANS Data Station Social Sciences and Humanities Collection.
Kuiper, G. from the Vrije Universiteit Amsterdam collected survey data on participants of a 1965 correspondence course in bookkeeping. The dataset includes motives for dropping out, attitudes towards study, motivation to start, and background variables like occupation, education, and religion. It was last updated on the DANS Data Station Social Sciences and Humanities platform in October 2025.