@article{seeuposequencelevelagenticrlwithconverge, title = {SeeUPO: Sequence-Level Agentic-RL with Convergence Guarantees}, author = {Tianyi Hu and Qingxu Fu and Yanxi Chen and Zhaoyang Liu and Bolin Ding}, year = {2026}, eprint = {2602.06554}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2602.06554}, }