Viverun/jits-legal-dataset
JITS Legal Dataset The only major open Indian legal dataset with citation-graph extraction, statutory section tagging, and IPC/CrPC→BNS/BNSS transition mapping — verified against the schemas of every comparable open dataset in this space (including one at 17.1M rows), which provide raw text and/or task labels but not structured legal extraction. Overview Disclaimer: This dataset is independently created for research and engineering use. It is not an official… See the full description on the dataset page: https://huggingface.co/datasets/Viverun/jits-legal-dataset.
v1.14.2: remove unsupported dataset loading script
Remove unsupported dataset loading script (HF datasets no longer supports repo-hosted scripts at all)
Update README.md
v1.14.1: HF-safe uniform structs - fix parquet DatasetGenerationError
v1.10: full corpus rebuild - 2,446 new judgments, two ReDoS fixes, case_number cleanup
v1.10: full corpus rebuild - 2,446 new judgments, two ReDoS fixes, case_number cleanup
Drop competitor names from headline claim
Headline verified structured-extraction edge vs. competitors
Remove 846 legacy per-document data/*.json files; add explicit dataset config
v1.10: full corpus rebuild - 2,446 new judgments, two ReDoS fixes, case_number cleanup
v1.10: full corpus rebuild - 2,446 new judgments, two ReDoS fixes, case_number cleanup
v1.10: full corpus rebuild - 2,446 new judgments, two ReDoS fixes, case_number cleanup
v1.8: CWP court fallback + refreshed release
v1.8: CWP court fallback + refreshed release
docs: update v1.7 post-rebuild metrics and integrity status
docs: update dataset card to v1.6 metrics and changelog
v1.5: Court codes corrected, tribunal IDs added, metadata extraction tightened
v1.5: Court codes corrected, tribunal IDs added, metadata extraction tightened
v1.5: Court codes corrected, tribunal IDs added, metadata extraction tightened
v1.4: Fix classification, transition temporal guardrails, and similarity signals
v1.4: Fix classification, transition temporal guardrails, and similarity signals
Docs: Add community feedback note
Docs: Add community feedback note
Fix: Upload normalized train.jsonl for HF Viewer support
Docs: Final Professional Polish (Disclaimer, Tags, Verification)
Docs: Add Schema Overview and JSON Example
Documentation: Remove v2.0 versions references
Fix: Correct metrics to verified counts (4,233 citations)
Contextualize metric notes for Hugging Face
Sync README.md and DATASET_CARD.md with correct metrics (846)
Fix: Correct judgment count to 846
Update dataset card to v1.3
Fix: Convert data to JSONL format for Viewer compatibility
Initial dataset upload (v2.0)
initial commit