Reasoning Model Is Superior LLM-Judge, Yet Suffers from Biases
This work addresses the problem of improving judgment accuracy and reducing biases in LLM-based evaluation systems, which is incremental as it builds on existing models with a new strategy.
The paper investigates whether Large Reasoning Models (LRMs) are better judges than non-reasoning LLMs, finding that LRMs outperform in accuracy, instruction-following, and robustness but still have biases, and proposes PlanJudge to reduce these biases.
This paper presents the first systematic comparison investigating whether Large Reasoning Models (LRMs) are superior judge to non-reasoning LLMs. Our empirical analysis yields four key findings: 1) LRMs outperform non-reasoning LLMs in terms of judgment accuracy, particularly on reasoning-intensive tasks; 2) LRMs demonstrate superior instruction-following capabilities in evaluation contexts; 3) LRMs exhibit enhanced robustness against adversarial attacks targeting judgment tasks; 4) However, LRMs still exhibit strong biases in superficial quality. To improve the robustness against biases, we propose PlanJudge, an evaluation strategy that prompts the model to generate an explicit evaluation plan before execution. Despite its simplicity, our experiments demonstrate that PlanJudge significantly mitigates biases in both LRMs and standard LLMs.