Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena
Zheng et al., 2023, Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena (Advances in Neural Information Processing Systems 36 (NeurIPS 2023), Datasets and Benchmarks Track)
Subscriber deck — free during early access
1 cards in “3.4 Addressing limitations” · whole deck
3.4 Addressing limitations
- How can position bias and math-grading errors in LLM judges be mitigated?Call the judge twice with the answers swapped and declare a win only if both orders agree (otherwise a tie). For math, have the judge answer the question independently first and insert that answer as a reference in the judging prompt, which cut the failure rate from 70% to 15%. Few-shot examples raised GPT-4’s consistency from 65% to 77.5%.
Swapping the answers’ order and requiring agreement neutralizes position bias - Zheng et al., 2023, Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena (Advances in Neural Information Processing Systems 36 (NeurIPS 2023), Datasets and Benchmarks Track), §3.4
Source
- Lianmin Zheng, Wei-Lin Chiang, Ying Sheng, Siyuan Zhuang, Zhanghao Wu, Yonghao Zhuang, Zi Lin, Zhuohan Li, Dacheng Li, Eric P. Xing, Hao Zhang, Joseph E. Gonzalez and Ion Stoica. 2023. Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena. In Advances in Neural Information Processing Systems 36 (NeurIPS 2023), Datasets and Benchmarks Track. https://papers.nips.cc/paper_files/paper/2023/hash/91f18a1287b398d378ef22505bf41832-Abstract-Datasets_and_Benchmarks.html