ToolSelf: Unifying Task Execution and Self-Reconfiguration via Tool-Driven Intrinsic Adaptation

TL;DR

ToolSelf introduces a unified framework for agent self-reconfiguration driven by tools, enabling autonomous adaptation during task execution, which significantly improves performance and generalization in complex, long-horizon tasks.

Contribution

The paper presents ToolSelf, a novel paradigm that integrates task execution and self-reconfiguration via tool-driven updates, and introduces CAT training for autonomous agent adaptation.

Findings

Achieves 24.1% average performance gain over baselines.

Rivals specialized workflows while maintaining generalization.

Enables agents to autonomously update goals and strategies during tasks.

Abstract

Agentic systems powered by Large Language Models (LLMs) have demonstrated remarkable potential in tackling complex, long-horizon tasks. However, their efficacy is fundamentally constrained by static configurations governing agent behaviors, which are fixed prior to execution and fail to adapt to evolving task dynamics. Existing approaches, relying on manual orchestration or heuristic-based patches, often struggle with poor generalization and fragmented optimization. To transcend these limitations, we propose ToolSelf, a novel paradigm enabling tool-driven runtime self-reconfiguration. By abstracting configuration updates as a callable tool, ToolSelf unifies task execution and self-adjustment into a single action space, achieving a phase transition from external rules to intrinsic parameters. Agents can thereby autonomously update their sub-goals and context based on task progression, and correspondingly adapt their strategy and toolbox, transforming from passive executors into dual managers of both task and self. We further devise Configuration-Aware Two-stage Training (CAT), combining rejection sampling fine-tuning with trajectory-level reinforcement learning to internalize this meta-capability. Extensive experiments across diverse benchmarks demonstrate that ToolSelf rivals specialized workflows while generalizing to novel tasks, achieving a 24.1% average performance gain and illuminating a path toward truly self-adaptive agents.