StepTool: Enhancing Multi-Step Tool Usage in LLMs via Step-Grained Reinforcement Learning.

@inproceedings{yu-2025-steptool, author = {Yuanqing Yu and Zhefan Wang and Weizhi Ma and Shuai Wang and Chuhan Wu and Zhiqiang Guo and Min Zhang}, editor = {Meeyoung Cha and Carl Yang and Senjuti Basu Roy and Chanyoung Park and Zhenhui Jessie Li and Noseong Park and Carl Yang and Senjuti Basu Roy and Zhenhui Jessie Li and Jaap Kamps and Kijung Shin and Bryan Hooi and Lifang He}, title = {{StepTool: Enhancing Multi-Step Tool Usage in LLMs via Step-Grained Reinforcement Learning}}, booktitle = {{Proceedings of the 34th ACM International Conference on Information and Knowledge Management, CIKM 2025, Seoul, Republic of Korea, November 10-14, 2025}}, pages = {3952--3962}, publisher = {ACM}, year = {2025}, url = {https://doi.org/10.1145/3746252.3761391}, doi = {https://doi.org/10.1145/3746252.3761391} }