Hugging Peers

Negative Advantage Is a Double-Edged Sword: Calibrating advantages in GRPO for Search Agents

☆ Save
Wu, Jiayi,Xie, Ruobing,Huang, Zeqian,Jiang, Lei,Xu, Can,Luo, Kangyang,Lin, Bochen,Gao, Ming
+1 more authors show fewer
Li, Xiang
Findings of the Association for Computational Linguistics: EMNLP 2026
published

Abstract

No abstract available.

Cite this paper

@article{jiayi2026,
  title = {Negative Advantage Is a Double-Edged Sword: Calibrating advantages in GRPO for Search Agents},
  author = {Wu, Jiayi and Xie, Ruobing and Huang, Zeqian and Jiang, Lei and Xu, Can and Luo, Kangyang and Lin, Bochen and Gao, Ming and Li, Xiang},
  journal = {Findings of the Association for Computational Linguistics: EMNLP 2026},
  year = {2026},
  month = {Jan},
}