论文54

华为论文提出 EvoTrias:将搜索规划、代码编写与评分分给三个角色

原标题:New Huawei paper finds that code-evolution agents do better and spend less when search planning, code writing, and scoring are handled by...

Rohan Paul@rohanpaul_ai发布时间 (北京时间)

华为一篇新论文提出 EvoTrias,把代码进化智能体的搜索规划、代码编写和评分分别交给控制器、编码智能体和智能体无法读取的评估器。在 7 个数学任务上它领先 6 个,单次运行 token 成本比 4 智能体的 CORAL 降低 64%;在 68 万行 C++ 仓库 OR-Tools 的 300 个实例上,求解质量提升 13.2%,而同一智能体单次运行平均为 -0.1%。

中文译文尚未完成,先为你展示原文。

New Huawei paper finds that code-evolution agents do better and spend less when search planning, code writing, and scoring are handled by 3 separate owners.

AlphaEvolve-style systems improve programs by writing code, scoring it, and picking what to try next. Some give the code writer little freedom, others let 1 coding agent steer the search at high token cost, and none hides the scorer.

EvoTrias gives each job its own owner. A controller steers the search, a coding agent like Claude Code writes code, and an evaluator the agent cannot read scores it.

On 7 math tasks it led 6 and cut token cost per run by 64% versus 4-agent CORAL. On OR-Tools, a 680K-line C++ repository, it lifted solver quality 13.2% on 300 instances, while 1-shot runs of the same agent averaged -0.1%.

For code-optimizing agent loops, hide the scorer from the agent and keep search planning out of the coding agent.

Agent 智能体AI 编码数据与训练论文研究#论文/研究#Agent#编码#数据/训练