Synthetic document classification dataset
A set of 248 corporate document images for training classifiers.
- Image classification
- Text classification
- English
Hugging Face datasets with plain-language descriptions: what's inside and what models to train on them. 4,788 in the catalog so far, with new ones added every hour.
Answered by Gemini 4 · the chat is kept only in this tab
A set of 248 corporate document images for training classifiers.
276k instruction-response pairs and multi-turn dialogues in French for language model training.
A corpus of scientific conference reviews with LLM-extracted claims and metrics.
Dataset for evaluating the safety of both final answers and intermediate reasoning steps of models.
Over 1.5 million ratings from real people for images generated from English text prompts.
A large corpus of English developer questions for training ticket priority classifiers.
English benchmark with a reproducible environment for evaluating task routing between AI agents.
US companies 10-K annual reports, split into sections and sentences, with sentiment labels.
Large cleaned corpus of real Russian-language reviews from 2023 for NLP tasks.
English human-bot conversations annotated with quality ratings for individual turns and full dialogues.
A set of paired English text continuations from humans and neural networks for style study and AI generation detector training.
Automatically annotated texts in 40 languages for training NER models.
A collection of English prompts for generating realistic photo portraits using the FLUX.1 model.
Annual MEDLINE/PubMed citation baseline in English XML with daily updates.
Around 3 million real-user texts across 55 languages annotated for toxicity to train and evaluate moderation filters.
Around 37k annotated examples of potential idioms in figurative or literal contexts.
Human-written Thai instruction dataset for training and evaluating models with local cultural awareness.
130 thousand examples of response source verification with annotation of hallucination types and errors.
A corpus of 500,000 multilingual texts linked to geographic coordinates across 123 regions.
NVIDIA dataset for training and evaluating Japanese personally identifiable information (PII) detection and masking models.
Records from ClinicalTrials.gov with metadata and pre-computed embeddings for biomedical tasks.
Cleaned version of Huatuo26M for training Chinese medical models.
Tabular-text security assessment dataset for public OpenClaw skills with scanner verdicts.
A dataset of English-language texts for studying discussions on Pill Ideologies subreddits.