讓 AI 助理自己改程序式時,怎麼安排搜尋、寫程式和評分,會影響成果與成本。
- Huawei 的論文提出 EvoTrias,讓控制器規劃搜尋、Claude Code 這類 AI 助理負責寫程式,再由 AI 助理看不到的評分器檢查結果。7 項數學任務中,EvoTrias 有 6 項表現最好,單次執行的 token 成本比 4-agent CORAL 少 64%。在有 680K 行 C++ 程式碼的 OR-Tools 專案中,它讓 300 個測試案例的求解器品質提升 13.2%。
- 同一個 AI 助理只執行一次,平均則下降 0.1%。
AI 伴隨問答
針對這篇報導有疑問?可直接點選快捷問題或向 AI 追問背景脈絡與技術細節:
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.