Archived records

Exact versions, evidence traces and recorded status.

These pilot records include general AI research used to develop the archive. Browse the legal AI paper index.

Refine resultsSubject, type, AI role, status & order

1–2 of 2 works

Research paper · Artificial intelligence

Meta-Harness: End-to-End Optimization of Model Harnesses

Yoonho Lee, Roshen Nair, Qizheng Zhang, and 3 others

Introduces an outer-loop coding agent that optimizes executable LLM harnesses by inspecting the full code, scores, and traces of prior candidates; one evaluation uses LawBench criminal-charge classification.

Limitation Legal evidence is limited to one LawBench criminal-charge classification benchmark within a broader three-domain AI-systems evaluation, and LegalArxiv has not independently reproduced any reported result.

Version & research context
Evaluation
Source-reported · not independently reproduced
Record status
External preprint · COLM 2026 accepted
Scope
Not jurisdiction-specific · EN
Contents
1 versions · 10 sections · 1 claim traces
Latest change
Initial public arXiv release.

Research collection · Human–AI research

Accelerating Scientific Research with Gemini: Case Studies and Common Techniques

David P. Woodruff, Vincent Cohen-Addad, Lalit Jain, and 33 others

A selected collection reporting human–AI collaboration across theoretical computer science, economics, optimization, and physics.

Limitation The cases are selected; the full denominator of projects, prompts, branches, and abandoned investigations is unavailable.

Version & research context
Evaluation
Claim-specific
Record status
Pilot reconstruction
Scope
Not applicable · EN
Contents
1 versions · 17 sections · 1 claim traces
Latest change
Color-description cleanup, a cosmic-string follow-up citation, and minor wording edits.

Print editions · Machine-readable access

Research updates

A monthly digest of selected legal AI papers, datasets and benchmarks, plus published research checks.