Hooshware

ESTABLISHING NEURAL LINK...

Back to News Radar

ResearchOct 15, 2026

Similar Models, Broken Odds: Where Luce's Axiom Fails in Chatbot Arena

Analyzing 1.8 million Chatbot Arena votes, researchers show that Luce's choice axiom fails for similar language models, as near-copies and related models…

arXiv – Computation and Language Read original report

Research

#Chatbot Arena#Bradley-Terry model#Choice Theory#Language Models#Evaluation#LLM#Release

Observed across 1 source

1 editorial report · 0 verified social mentions. The most authoritative report leads while later evidence completes the story.

arXiv – Computation and LanguagePrimary source

Similar Models, Broken Odds: Where Luce's Axiom Fails in Chatbot Arena

Open report

BTC—

SYNCING

ETH—

SYNCING

NVDA—

SYNCING

MSFT—

SYNCING

Customize