Datasets:
Modalities:
Text
Formats:
json
Size:
< 1K
ArXiv:
Tags:
benchmark
long-context
multi-hop-reasoning
source-internal-reasoning
evidence-withheld
document-qa
License:
Commit History
-
Fix evaluation harness and normalize rubrics
Verified signatureNormalize rubric encodings for wildtrace-0195, wildtrace-0371, and wildtrace-0419; support HF JSONL ground-truth fields; create output directories; enforce complete judge panels; report Scored and All481 aggregation; refresh documentation and checksums.b37dd4f
-
Add Gemini-native judge compatibility to eval harness
Verified signature9c7ab6a -
Clean license and corpus header metadata in WildTrace strict481
Verified signatureSet mixed-license dataset card metadata, clarify annotation/source text license scope, neutralize construction-only corpus headers while preserving offsets, and refresh manifests/checksums.bb74a71 -
Clean WildTrace strict481 public release
Verified signatureApply question-text metadata leak cleanup, remap public task/source ids, strip internal iteration metadata, remove stale leaderboard artifacts, and refresh checksums/manifests.fb2ffe9
-
Clarify license scope: CC-BY covers annotations/scripts only; source texts retain PD/§105/Project-Gutenberg terms + US-jurisdiction note
Verified signaturee6091d6 -
Add per-source provenance manifest (PD/§105/PG basis per corpus file)
Verified signature23d1ed1 -
Release WildTrace strict481 public package
Verified signature950f6b2 -
initial commit
Verified signature50bacf3