Progra: Progress-Aware Reinforcement Learning for Multi-Turn Function Calling

Huacan Chai, Zijie Cao, Maolin Ran, Yingxuan Yang, Jianghao Lin, pengxin, HairuiWang, Renjie Ding, Ziyu Wan, Muning Wen, Weiwen Liu, Weinan Zhang, fei huang, Ying Wen

Keywords: Function Calling, Large Language Model

Abstract: Large language models (LLMs) have achieved impressive success in single-turn function calling, yet real-world applications such as travel planning or multi-stage data analysis typically unfold across multi-turn conversations. In these settings, models must not only issue accurate function calls at each step but also maintain progress awareness, the ability to summarize past interactions and plan future actions to ensure coherent, long-horizon task execution. Existing approaches, however, either reduce multi-turn training to isolated single-turn samples, which neglects task-level planning, or employ end-to-end reinforcement learning (RL) that struggles with redundancy and lacks explicit integration of progress awareness. To overcome these limitations, we introduce Progra, a framework that explicitly incorporates progress awareness into LLM training for multi-turn function calling. Progra combines (i) a Progress Awareness Generation (PAG) pipeline, which automatically constructs datasets coupling conversation summaries with future task planning, and (ii) a Progress Awareness-Guided Reinforcement Learning (PAG-RL) algorithm, which integrates progress awareness into RL training to reduce contextual redundancy and improve alignment between local actions and global task completion. Empirical results on two public benchmarks demonstrate that Progra significantly outperforms existing methods, highlighting the effectiveness of progress awareness in enabling robust and efficient multi-turn function calling. Our code can be available at https://anonymous.4open.science/r/Progra_ICLR26-57F0

Primary Area: foundation or frontier models, including LLMs

Submission Number: 11738