LoopTool: Closing the Data-Training Loop for Robust LLM Tool Calls

TL;DR

LoopTool introduces a fully automated, closed-loop framework that iteratively refines data and models for robust tool-using LLMs, significantly improving performance without relying on costly external APIs.

Contribution

We propose LoopTool, a novel framework that integrates data synthesis and model training in a closed loop, enhancing LLM tool use capabilities through self-refinement.

Findings

Model trained with LoopTool outperforms larger models on benchmarks.

Closed-loop data refinement improves model accuracy and robustness.

Eliminates dependence on expensive external APIs.

Abstract

Augmenting Large Language Models (LLMs) with external tools enables them to execute complex, multi-step tasks. However, tool learning is hampered by the static synthetic data pipelines where data generation and model training are executed as two separate, non-interactive processes. This approach fails to adaptively focus on a model's specific weaknesses and allows noisy labels to persist, degrading training efficiency. We introduce LoopTool, a fully automated, model-aware data evolution framework that closes this loop by tightly integrating data synthesis and model training. LoopTool iteratively refines both the data and the model through three synergistic modules: (1) Greedy Capability Probing (GCP) diagnoses the model's mastered and failed capabilities; (2) Judgement-Guided Label Verification (JGLV) uses an open-source judge model to find and correct annotation errors, progressively purifying the dataset; and (3) Error-Driven Data Expansion (EDDE) generates new, challenging samples based on identified failures. This closed-loop process operates within a cost-effective, open-source ecosystem, eliminating dependence on expensive closed-source APIs. Experiments show that our 8B model trained with LoopTool significantly surpasses its 32B data generator and achieves new state-of-the-art results on the BFCL-v3 and ACEBench benchmarks for its scale. Our work demonstrates that closed-loop, self-refining data pipelines can dramatically enhance the tool-use capabilities of LLMs.