datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
RefWave-Cluster-Runsevalarc-casebook
EvalArc Casebook
The same 93.75% score can pass one acceptance gate and fail another.
Inspect the rules, actual failed checks and original Docker records in a
filterable table. This is the data companion to the
interactive evidence lab.
In the default suite_jobs view, compare support-partial and
support-protected. Both use the same frozen defective policy, score 93.75%
and fully resolve 0/2 attempts. The deliberately permissive rule accepts partial
progress; the rule requiring… See the full description on the dataset page: https://huggingface.co/datasets/glayguo/evalarc-casebook.mining-legal-arguments-us-corporate-case-law
Mining Legal Arguments in U.S. Corporate Case Law
This dataset contains span-level functional labels and directed support relations for 42 U.S. federal tax opinions concerning corporate reorganizations under I.R.C. Section 368. The opinions range in citation year from 1935 to 1987. Two law students annotated the cases, and a law professor adjudicated the final case-level representations. Ten cases also include the two independent annotations used for inter-annotator agreement… See the full description on the dataset page: https://huggingface.co/datasets/lbrenap1/mining-legal-arguments-us-corporate-case-law.cross-unlearning-case-400Supreme-Court-Cases-1830-2019
US Supreme Court Legal Corpus (1830–2019)
Overview
A comprehensive, production-ready AI training dataset containing 456,589 documents from 122,930 US Supreme Court cases spanning 190 years (1830–2019).
This corpus captures the full adversarial record — petitions for certiorari, respondent briefs, reply briefs, amicus curiae filings, appendices, oral argument transcripts, and opinions. It is one of the most complete collections of Supreme Court procedural and… See the full description on the dataset page: https://huggingface.co/datasets/OwnedByDanes/Supreme-Court-Cases-1830-2019.Sherlock-Case-Files
Sherlock Case Files 📁
Sherlock Case Files is a synthetic multilingual dataset for schema-guided
information extraction. Each case asks a model to read a compact JSON schema and a text, then return exactly one JSON object matching that schema.
The dataset covers short snippets and long documents across varied domains and formats. It includes distractors and missing fields, represented by null, in English, Italian, Spanish, French, Portuguese, and German. Metadata supports… See the full description on the dataset page: https://huggingface.co/datasets/derogab/Sherlock-Case-Files.ambiguity-casebook
Dual-Use Ambiguity Casebook
A 35-row, single-annotator research corpus for studying context-sensitive
adjudication in AI-mediated biology. Records follow one exact 21-field schema
and cover six descriptive categories.
This is a dataset release, not a model leaderboard, compliance tool, or
laboratory guide. Raw model responses, model-comparison results, live-provider
evaluation code, and adversarial failure-mode material are intentionally
excluded.
Dataset structure… See the full description on the dataset page: https://huggingface.co/datasets/jang1563/ambiguity-casebook.bert-base-multilingual-cased-toksuite-detokenizedTraining data of the model detokenized in the exact order seen by the model.
The training data is partitioned into 8 chunks (chunk-0 through chunk-7), based on the GPU rank that generated the data. Each chunk contains detokenized text files in JSON Lines format (.jsonl).
Caselaw_Access_Project_FAISS_index
The Caselaw Access Project
In collaboration with Ravel Law, Harvard Law Library digitized over 40 million U.S. court decisions consisting of 6.7 million cases from the last 360 years into a dataset that is widely accessible to use. Access a bulk download of the data through the Caselaw Access Project API (CAPAPI): https://case.law/caselaw/
Find more information about accessing state and federal written court decisions of common law through the bulk data service documentation here:… See the full description on the dataset page: https://huggingface.co/datasets/free-law/Caselaw_Access_Project_FAISS_index.aria-soc-cases
AriaSOC laboratory cases
Synthetic SIEM alerts and case labels used by AriaSOC.
Collection: Aria AI — Cybersecurity.
Live Gradio hosting follows the personal-PRO / organization-card pattern used by the Aria AI demos.
This is generated data (seed 24). It is for reproducible demos and tests, not a customer extract. Addresses are documentation-range or RFC1918 laboratory values. Tokens and mailboxes are fake and redacted in the demo.
Files
File
Grain
Notes… See the full description on the dataset page: https://huggingface.co/datasets/AriaAICompany/aria-soc-cases.california_tos_court_cases_32k_v1construction-accident-cases-weather
건설현장 사고사례 + 날씨 데이터셋 / Construction Accident Cases with Weather
과거 건설현장 사고사례(현장·공종·작업·피해·사고유형 등) 레코드에
발생 시점의 기상 정보(기온·체감온도·풍속·습도) 를 결합한 한국어 데이터셋입니다.
사고-기상 상관관계 분석, 위험요인 모델링, 건설안전 분석용 LLM/ML 학습의 원천 데이터로 활용할 수 있습니다.
A Korean dataset that joins construction-site accident cases (site, work type, task, damage,
accident type, etc.) with the weather conditions at the time of the accident (temperature,
apparent temperature, wind speed, humidity). Useful for accident–weather correlation… See the full description on the dataset page: https://huggingface.co/datasets/swordKoala/construction-accident-cases-weather.case0-traces
Case Zero - agent traces
Real traces from Case Zero,
a procedural detective game where a single Qwen2.5-1.5B model (in-process llama.cpp,
CPU-only, no cloud APIs) authors a complete mystery and then role-plays every suspect
live under interrogation.
case0_traces.jsonl - one JSON object per line:
type: "generation_call" - one pipeline LLM call while authoring a case: the exact
prompt, the raw completion, sampling params, and latency. Two calls author a full
case (world+cast… See the full description on the dataset page: https://huggingface.co/datasets/build-small-hackathon/case0-traces.japanese-legal-cases-2025california_tos_court_cases_v1caselawqa_leaderboard_requests