arXiv cs.AI· Meihong Jia, Yasemin Timar, Jonathan Taws·· 13 小时前AI 评分31
GuideEvolve：结合专家指导与 3Ts 机制的自进化智能体框架
GuideEvolve: Self-Evolving Agent Framework with Expert Guidance and 3Ts Mechanisms
AI 导读
GuideEvolve 是一个结合专家指导与 3Ts 机制（Think Fast and Slow、Throwback、Termination by choice）的自进化智能体框架。
正文 · 原文
Abstract:This study introduces GuideEvolve, a self-evolving agent framework incorporating expert guidance and three autonomous regulatory mechanisms, collectively termed the 3Ts: Think Fast and Slow, Throwback, and Termination by choice. The framework addresses critical limitations in extant self-evolving agent paradigms through dual innovations: (1) a task-specific expert-evaluation-driven evolutionary pipeline, and (2) the integration of 3Ts mechanisms to facilitate cognitive optimization, performance recovery via temporal rollback, and adaptive termination protocols. Empirical validation across the MGSM and MMLU benchmarks demonstrates that GuideEvolve attains performance metrics comparable to current state-of-the-art models (89.8% accuracy versus Gödel Agent's 90.6% on MGSM; 89.1% versus 87.9% on MMLU), while exhibiting notably reduced variance (25-59%) and enhanced robustness. Furthermore, the framework achieves a 100% evolution success rate, superior to the baseline's 82%, diminishes required optimization iterations by approximately 50% (from 30 iterations in baseline to 15.1 and 13.4 across tasks), and provides improved stability for production deployment.
| Comments: | 14 pages, 10 images/figures |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.30017 [cs.AI] |
| (or arXiv:2610.30017v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.30017 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yasemin Timar [view email]
[v1]
Wed, 21 Oct 2026 16:46:22 UTC (3,588 KB)
来源：arXiv cs.AI · arxiv.org