用自然语言控制程序演化，让修改更精准高效。
ELMER: Evolutionary Language Model that Explores and Refines
- 以自然语言描述为搜索空间，通过条件指令控制变异强度
- 在252次搜索中提升行为校准度与效率，语言描述表现最优
- 适合需要可控程序生成的开发与自动化测试场景
程序演化可判断突变是否有益，但很少能控制其在行为空间中的移动范围。语法编辑大小是不可靠的代理：小改动可能改变所有行为，而大重构却可能保留相同执行轨迹。我们提出进化语言模型ELMER，可在自然语言策略描述中搜索，并编译出带类型的可执行程序。一个全微调的Qwen3-8B模型学习三种任务相关操作：条件语义突变、自然语言到领域特定语言（GPTL）编译、以及GPTL到自然语言翻译。该模型通过直接偏好优化（oDPO）进行微调，输入条件为突变强度（低、中、高）。在252次固定预算的进化搜索中，oDPO提升了行为校准度和有限预算下的搜索效率。自然语言描述达到最高观测留出适应度。分析表明，条件输入（突变强度）系统性地改变了语义编辑组成，且语言突变在匹配的小到中等行为偏移下能更好地保留父代适应度。结果表明，语言可作为可调控、基于执行的可执行程序空间搜索表示。
原文摘要 · Abstract (English)
Program evolution can measure whether a mutation helped, but it rarely controls how far the mutation moves in behavior space. Syntactic edit size is an unreliable proxy: a small code change can alter nearly every action, while a larger rewrite can preserve the same execution trace. We introduce an Evolutionary Language Model that searches over natural-language policy descriptions and compiles typed programs for execution. A fully fine-tuned Qwen3-8B model learns three task-conditioned operations: conditional semantic mutation, natural language to domain-specific language (GPTL) compilation, and GPTL to natural language translation. The model is fine-tuned with conditional input on the mutation strength (low, medium, high) using Direct Preference Optimization (oDPO). Across 252 fixed-budget evolutionary searches, oDPO improves both behavioral calibration and finite-budget search efficiency. Natural-language attains the highest observed held-out fitness. Our analysis shows that the condition input (mutation strength) systematically changes semantic edit composition and that language mutations preserve more parent fitness at matched small-to-moderate behavioral displacement. These results show that language can serve as a steerable, execution-grounded search representation over executable program space.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。