English

Incentivizing Agentic Reasoning in LLM Judges via Tool-Integrated Reinforcement Learning

Computation and Language 2026-02-24 v2 Artificial Intelligence Machine Learning

Abstract

Large Language Models (LLMs) are widely used as judges to evaluate response quality, providing a scalable alternative to human evaluation. However, most LLM judges operate solely on intrinsic text-based reasoning, limiting their ability to verify complex constraints or perform accurate computation. Motivated by the success of tool-integrated reasoning (TIR) in numerous tasks, we propose TIR-Judge, an end-to-end RL framework for training LLM judges that integrates a code executor for precise evaluation. TIR-Judge is built on three principles: (i) diverse training across verifiable and non-verifiable domains, (ii) flexible judgment formats (pointwise, pairwise, listwise), and (iii) iterative RL that bootstraps directly from the initial model without distillation. On seven public benchmarks, TIR-Judge surpasses strong reasoning-based judges by up to 6.4% (pointwise) and 7.7% (pairwise), and achieves listwise performance comparable to Claude-Opus-4 despite having only 8B parameters. Remarkably, TIR-Judge-Zero - trained entirely without distilled judge trajectories, matches the performance of distilled variants, demonstrating that tool-augmented judges can self-evolve through iterative reinforcement learning.

Keywords

Cite

@article{arxiv.2510.23038,
  title  = {Incentivizing Agentic Reasoning in LLM Judges via Tool-Integrated Reinforcement Learning},
  author = {Ran Xu and Jingjing Chen and Jiayu Ye and Yu Wu and Jun Yan and Carl Yang and Hongkun Yu},
  journal= {arXiv preprint arXiv:2510.23038},
  year   = {2026}
}

Comments

ICLR 2026

R2 v1 2026-07-01T07:07:11.350Z