@article{eapoentropydrivenadaptivepositivenegativ, title = {EAPO: Entropy-Driven Adaptive Positive-Negative Sample Weighting for Policy Optimization in Open-Ended QA}, author = {Yunsheng Zeng and Gen Li and Yuwei Miao and Xiandong Li and Yujin Wang and Siyu Chen and Luning Wang and Yunhao Qiao and Junfeng Wang and Jianwei Lv and Bo Yuan}, year = {2026}, eprint = {2605.27846}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2605.27846}, }