Self-Guided Function Calling in Large Language Models via Stepwise Experience Recall
This addresses inefficiencies in function calling for LLMs, offering a scalable solution for developers and researchers, though it is incremental over existing retrieval-based methods.
The paper tackles the problem of LLMs struggling with multi-step tool usage by proposing SEER, a self-guided method that retrieves stepwise experiences from an updated pool, achieving accuracy improvements of 6.1% on easy and 4.7% on hard questions in ToolQA, and up to 23.38% on real-world benchmarks.
Function calling enables large language models (LLMs) to interact with external systems by leveraging tools and APIs. When faced with multi-step tool usage, LLMs still struggle with tool selection, parameter generation, and tool-chain planning. Existing methods typically rely on manually designing task-specific demonstrations, or retrieving from a curated library. These approaches demand substantial expert effort and prompt engineering becomes increasingly complex and inefficient as tool diversity and task difficulty scale. To address these challenges, we propose a self-guided method, Stepwise Experience Recall (SEER), which performs fine-grained, stepwise retrieval from a continually updated experience pool. Instead of relying on static or manually curated library, SEER incrementally augments the experience pool with past successful trajectories, enabling continuous expansion of the pool and improved model performance over time. Evaluated on the ToolQA benchmark, SEER achieves an average improvement of 6.1% on easy and 4.7% on hard questions. We further test SEER on $τ$-bench, which includes two real-world domains. Powered by Qwen2.5-7B and Qwen2.5-72B models, SEER demonstrates substantial accuracy gains of 7.44% and 23.38%, respectively.