Reproduce “Legal Guidance in the Wild: How Users Seek Legal Help in Real-World LLM Conversations” 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
Amirreza Mohseni, Gerasimos Spanakis
The authors develop a hierarchical classifier to distinguish legal-content conversations from requests for practical legal guidance, then classify guidance-positive conversations by legal topic. In an enriched sample of 2,122 WildChat conversations, 972 of 1,367 legal-content conversations seek guidance (71.1%). ModernBERT-large reaches .671 topic macro-F1 versus .559 for a lexical baseline across 12 represented topics; the difference is not statistically significant.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.