arXiv cs.AI· Zhengyang Ai, Tong Liu, Jingxian Tang, Xiaobing Du, Tao Xiao, Fengshi Zhang, Zishuo Li, Xiaodong Ai, Pinyan Lu·· 1 小时前AI 评分46
EvoTrias：面向智能体程序演化的三权分立架构
Where Autonomy Belongs: A Separation of Powers for Agentic Program Evolution
AI 导读
arXiv 论文提出 EvoTrias，一种将 LLM 驱动的程序演化中 govern、propose、judge 三类决策分离到独立所有者的架构：Structured Governance、Open Proposal 与 Protected Judgment，各所有者仅通过 typed contracts 交互，并可作为插件把现有 coding agent 变成演化系统。
正文 · 原文
Abstract:LLM-driven program evolution rests on three decisions: govern directs the search from accumulated evidence, propose constructs the next candidate, and judge settles validity and fitness. All three appear in every system, and existing work splits into two lines over who makes each. Controller-centric systems strengthen govern and neglect the autonomy of propose, which caps what evolution can reach. Agent-centric systems let a coding agent take over both govern and propose, at a cognitive and computational cost. Neither line protects the independent authority of judge, so benchmark reliability suffers. These failures share one root: autonomy is misallocated across the three decisions. We address this problem with EvoTrias, a separation-of-powers architecture in which each decision has one owner: Structured Governance for what spans evolution rounds, Open Proposal for construction inside a task, and Protected Judgment behind a private evaluator. These owners exchange typed contracts only, and runtime hooks reinforce the boundaries. This separation also lets EvoTrias run as a plugin that turns an existing coding agent into an evolution system. EvoTrias sets a new state of the art on mathematical, systems, and algorithm-design benchmarks. Three further results show what each owner buys: separating Governance from Proposal cuts token cost by 64%, Open Proposal extends evolution to a 680K-line industrial repository, and Protected Judgment keeps benchmark scores faithful to transfer.
| Comments: | 58 pages, 26 figures |
| Subjects: | Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2611.15309 [cs.AI] |
| (or arXiv:2611.15309v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.15309 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Zhengyang' Ai [view email]
[v1]
Wed, 11 Nov 2026 10:44:21 UTC (1,796 KB)
来源：arXiv cs.AI · arxiv.org