CodeEvolve：用运行时验证驱动 LLM 进化式代码优化
Salesforce Research 出的代码优化框架，改代码前先跑测试再进搜索，Apex 测试通过数从 7.9 拉到 19.5，Java 还快了 15 倍。
Salesforce Research 发布 CodeEvolve 框架，把 MCTS 引入 LLM 进化式代码优化，候选代码必须通过构建、单元测试、性能检查、静态分析和 LLM 审查才能进入搜索种群。在 20 个 Apex 程序上，平均 19.5 个通过全部测试，OpenEvolve 基线为 7.9 个。有效程序数量的方差从 ±4.16 降到 ±0.93。对 7 个生产环境 Java 热点函数平均取得 15.22 倍加速，且行为保持正确。
𝗖𝗼𝗱𝗲𝗘𝘃𝗼𝗹𝘃𝗲: LLM-Driven Evolutionary Optimization with Runtime-Enriched Target Selection for Multi-Language Code Enhancement https://t.co/t2iNDsreuy
LLM code edits only help if they preserve behavior. CodeEvolve runs every candidate through builds, unit tests, performance checks, static analysis, and LLM review, and only validated variants enter the search population.
→ 19.5 of 20 Apex programs pass all tests on average with the full MCTS-augmented system, up from 7.9 for the OpenEvolve baseline → Variance in valid programs drops from ±4.16 to ±0.93 → 15.22× average speedup across 7 production Java hotspot functions, with correctness intact
Authors: Ajay Krishna Borra, Wenzhuo Yang, Samarth Arora, Akhilesh Deepak Gotmare @akhilesh_gotmare, Gokulakrishnan Gopalakrishnan, Tharun Gali, Madhav Rathi, Doyen Sahoo @doyensahoo, Manpreet Singh, Mayuresh Verma, Laksh Venka, Shuchita Singh
#FutureOfAI #EnterpriseAI #SoftwareEngineering
