Datasets:
Modalities:
Text
Formats:
parquet
Languages:
English
Size:
1K - 10K
Tags:
knowledge-integration
knowledge-editing
continual-learning
retrieval-augmented-generation
generalization
medical
License:
Commit History
-
Sync croissant.json with paper title and updated stats
- Update title in description and citeAs - Fix conflict-flagged count 154 -> 161 in rai:dataLimitations - Add missing locality-cascade fallback step to preprocessing protocol (12 steps now) - Switch license URL to canonical CC URL - Change creator @type from Person to Organization - Add per-column descriptions to all 26 fields876de36 -
Align title with paper; use precise label-distribution split
cac6b0b -
Sync README to canonical paper numbers
- Update H1 title to match paper - Update label distribution to 29 / 43 / 28 - Update conflict-flagged count to 161 (2.6%)6526ee5 -
Remove redundant croissant_rai.json (fields already merged into croissant.json)
Verified signature8ec1814 -
Refresh RAI fragment with full set of fields
Verified signature3cfb6b2 -
Full RAI + prov fields (rai:dataCollectionType, dataAnnotationProtocol, hasSyntheticData, prov:wasDerivedFrom, prov:wasGeneratedBy)
Verified signature9c7cd83 -
Add citeAs / datePublished / version to Croissant
Verified signature0902fed -
Refresh croissant.json with full schema + RAI
Verified signaturea9d77ac -
Add full Croissant metadata with RAI fields
Verified signature95fa629 -
Initial MedKIT release (anonymous NeurIPS submission)
Verified signatured864c99
-
Initial MedKIT release (anonymous NeurIPS submission)
Verified signaturef6165ab -
Initial MedKIT release (anonymous NeurIPS submission)
Verified signature98113f9 -
Initial MedKIT release (anonymous NeurIPS submission)
Verified signaturecb67c36 -
Initial MedKIT release (anonymous NeurIPS submission)
Verified signaturee029a72 -
initial commit
Verified signature9a23751