Sangdo Han
Titles as the sources gave them. In 1 entry, with 0 checked quotations.
In the Base
Every entry they appear in.
Swapping the Judge Changes the Ranking: A Preregistered Study of Cross-Provider Replication in LLM-Judged System Comparisons
According to the abstract, Sangdo Han and Moonsoo Kim tested whether repeating an AI-judged comparison with a judge from another provider is a reliable check. Five judge models re-judged 4,000 Chatbot Arena battles; agreement with human votes was low and rankings shifted with the judge.