04:00
2026-10-15
arxiv.org
large-language-models
Similar Models, Broken Odds: Where Luce's Axiom Fails in Chatbot Arena
A paper posted to arXiv (2610.19095v1) reports that Chatbot Arena's Bradley-Terry ranking model violates Luce's choice axiom, with votes between versions of the same model proving 1.37 times as decisi…