Crowd Comparative Reasoning: Unlocking Comprehensive Evaluations for LLM-as-a-Judge
This addresses the limitation of incomplete evaluations in LLM-as-a-Judge for researchers and practitioners, though it is incremental as it builds on existing auto-evaluation methods.
The paper tackles the problem of unreliable LLM-as-a-Judge evaluations by proposing Crowd-based Comparative Evaluation, which uses crowd responses to expose deeper details and guide more comprehensive chain-of-thought judgments, resulting in an average accuracy gain of 6.7% across five benchmarks.
LLM-as-a-Judge, which generates chain-of-thought (CoT) judgments, has become a widely adopted auto-evaluation method. However, its reliability is compromised by the CoT reasoning's inability to capture comprehensive and deeper details, often leading to incomplete outcomes. Existing methods mainly rely on majority voting or criteria expansion, which is insufficient to address the limitation in CoT. We propose Crowd-based Comparative Evaluation, which introduces additional crowd responses to compare with the candidate responses, thereby exposing deeper and more comprehensive details within the candidate responses. This process effectively guides LLM-as-a-Judge to provide a more detailed CoT judgment. Extensive experiments demonstrate that our approach enhances evaluation reliability, achieving an average accuracy gain of 6.7% across five benchmarks. Moreover, our method produces higher-quality CoTs that facilitate judge distillation and exhibit superior performance in rejection sampling for supervised fine-tuning (SFT), referred to as crowd rejection sampling, thereby enabling more efficient SFT. Our analysis confirms that CoTs generated by ours are more comprehensive and of higher quality, and evaluation accuracy improves as inference scales.