Skip to content

Latest commit

 

History

History
46 lines (34 loc) · 1.28 KB

File metadata and controls

46 lines (34 loc) · 1.28 KB

Annotation Protocol

Purpose

The human-validation protocol checks whether automated and model-judge scores align with human legal review, and whether the failure modes found in real cases differ from public-exam scoring.

Human-Scored Subset

Subset Items Model groups Human-validation responses
Public legal exams 80 20 1,600
Chinese real cases 10 20 200
Total 90 20 1,800

Review Focus

Human review checks:

  • source and reference-answer alignment for public-exam rows;
  • issue and stance consistency for real-case rows;
  • privacy and de-identification constraints;
  • citation relevance;
  • constraint extraction;
  • argument validity;
  • cases where automated scores appear too generous or too punitive.

Preferred Review Notes

Review notes should be brief and audit-friendly. Useful tags include:

  • opposite stance;
  • invented facts;
  • missing operative constraint;
  • weak citation linkage;
  • wrong legal domain;
  • missed key issue;
  • thin rule-to-fact analysis;
  • mostly aligned with minor gaps.

Reliability Reporting

A full public release should report reviewer training, double-coded subset size, agreement statistics, and adjudication procedure. The public repository does not include raw human review sheets.