arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2605.10366cs.AI
收藏

EGL-SCA:图推理代理中协同指令与工具的结构信用分配

EGL-SCA: Structural Credit Assignment for Co-Evolving Instructions and Tools in Graph Reasoning Agents

Harbin Institute of Technology(哈尔滨工业大学);Peng Cheng Laboratory(鹏城实验室);School of Computer Science and Technology(计算机科学与技术学院)

Zike Yuan, Yukun Cao, Han Zhang, Jianzhi Yan, Le Liu, Cai ke, Yue Yu, Hui Wang, Ming Liu, Bing Qin

AI总结:

本文提出EGL-SCA框架,通过结构信用分配机制协同优化指令与工具,提升图推理代理的结构正确性与执行效率,实验显示其在四个基准测试中达到92.0%的平均成功率。

摘要:

图推理代理需从自然语言输入中重构结构图实例,判断现有计算资源是否足够,并在严格执行协议下与工具交互,同时满足检查结构正确性的外部验证器。现有方法通常单独优化指令或工具侧,导致失败后更新不明确。本文提出EGL-SCA,一个以验证器为中心的双空间框架,通过指令侧策略空间和工具侧程序空间协同建模图推理代理。核心机制是结构信用分配,将轨迹证据映射到条件更新,精确将失败路由到提示优化或工具合成与修复。为提供双空间适应的充分学习信号,引入按任务家族分层的训练分布,结合帕累托式保留策略平衡成功、通用性和简洁性。实验表明,EGL-SCA在四个图推理基准测试中达到92.0%的平均成功率,通过有效协同优化指令与工具,显著优于纯提示和固定工具箱基线。

英文摘要:

Graph reasoning agents operating from natural-language inputs must solve a coupled problem: they must reconstruct a structured graph instance from text, decide whether existing computational assets are sufficient, interact with tools under a strict execution protocol, and satisfy an external verifier that checks structured correctness rather than textual plausibility. Existing approaches usually improve either the instruction side or the tool side in isolation, which leaves unclear what should be updated after failure. We propose EGL-SCA, a verifier-centric dual-space framework that models a graph reasoning agent using two collaborative components: an instruction-side policy space for reasoning strategies, and a tool-side program space for executable algorithmic tools. Our central mechanism is structural credit assignment, which maps trajectory evidence to conditional updates, precisely routing failures to either prompt optimization or tool synthesis and repair. To provide sufficient learning signals for dual-space adaptation, we introduce a training distribution stratified by task family, coupled with a Pareto-style retention strategy to balance success, generality, and parsimony. Experiments on four graph reasoning benchmarks show that EGL-SCA achieves a state-of-the-art 92.0\% average success rate. By effectively co-evolving instructions and tools, our framework significantly outperforms both pure-prompting and fixed-toolbox baselines.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑