Reproduce “Legal-R1: Agentic Retrieval for Evidence-Based Legal Reasoning” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Huaiwen Pang, Jiale Han, Junying Chen, Wenjun Xiao, Benyou Wang
Legal-R1 makes legal question answering iterative evidence gathering. A Lawyer model plans source-specific searches and decides when to stop, while a Judge writes answers from retrieved evidence. On benchmarks, its 8B system averages 60.28, versus 45.59 for 7B R-Search; the 284B model scores 66.47. More retrieval turns and tokens accompany the score; legal validity in practice remains unestablished.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.