Try, Check and Retry: A Divide-and-Conquer Framework for Boosting Long-context Tool-Calling Performance of LLMs

2026-03-11

AI Summary

Existing approaches struggle to effectively manage a large number of noisy candidate tools in long-context tool-augmented tasks, limiting the practical deployment of large language models. This work proposes Tool-DC, a novel framework that uniquely integrates divide-and-conquer strategy with self-reflection through an iterative “try–check–retry” paradigm to reduce reasoning complexity. Tool-DC features two variants: a training-free version (Tool-DC TF) enabling plug-and-play usability and a trainable version (Tool-DC TB) optimized for efficient inference. Experimental results demonstrate that Tool-DC (TF) achieves an average performance gain of 25.10% on BFCL and ACEBench benchmarks, while Tool-DC (TB) empowers Qwen2.5-7B to match or even surpass closed-source models such as OpenAI o3 and Claude-Haiku-4.5 in tool-calling capability.

Technology Category

• Natural Language Processing: (Large) Language Models
• Constraint Satisfaction and Optimization: Solvers and Tools
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Tool-calling empowers Large Language Models (LLMs) to interact with external environments. However, current methods often struggle to handle massive and noisy candidate tools in long-context tool-calling tasks, limiting their real-world application. To this end, we propose Tool-DC, a Divide-and-Conquer framework for boosting tool-calling performance of LLMs. The core of Tool-DC is to reduce the reasoning difficulty and make full use of self-reflection ability of LLMs via a "Try-Check-Retry" paradigm. Specifically, Tool-DC involves two variants: 1) the training-free Tool-DC (TF), which is plug-and-play and flexible; 2) the training-based Tool-DC (TB), which is more inference-efficient. Extensive experiments show that both Tool-DC methods outperform their counterparts by a clear margin. Tool-DC (TF) brings up to +25.10% average gains against the baseline on BFCL and ACEBench benchmarks, while Tool-DC (TB) enables Qwen2.5-7B to achieve comparable or even better performance than proprietary LLMs, e.g., OpenAI o3 and Claude-Haiku-4.5.

Problem

• tool-calling
• long-context
• noisy candidate tools
• LLMs
• real-world application

Innovation

• Divide-and-Conquer
• Tool Calling
• Long-context Reasoning
• Self-reflection
• Large Language Models