让材料搜索智能体学会从每步修改中吸取经验，提升多目标优化效率。
TRACE: Transition-Aware Residual Control for Multi-Objective Materials Discovery
- 以修改操作为反馈单元，记录修改前后的性能变化
- 在相同基底上将命中率从18.13%提升至25.96%
- 适合需要精细调控多属性材料的科研人员
使用大模型智能体进行多目标材料发现时，不仅受限于可生成的候选数量，更受限于每次昂贵的性能评估能否有效指导下一步搜索。现有智能体主要存储已评估材料及其得分，仅知道哪些材料成功，却无法判断哪些可执行的修改带来了有效的性质变化，导致在目标冲突时难以进行局部优化——一个修改可能提升某一性质，却损害另一性质。我们提出TRACE，一种面向转换的残差控制框架，将已评估的修改作为反馈的基本单元。TRACE记录每次局部优化为“父修改-子修改”转变及观察到的性质变化量，聚合转变证据以估计可复用的修改效果，并根据预测减少当前候选剩余约束违反的能力来排序未来修改，同时避免对已满足目标造成损害。在相同的基底对比实验中，TRACE相较当前最优的LLM代理基准LLEMA，将宏观平均命中率从18.13%提升至25.96%。
原文摘要 · Abstract (English)
Multi-objective materials discovery with LLM agents is often limited not only by how many candidates can be proposed, but by how effectively each costly property evaluation informs the next search step. Existing agents mainly store evaluated candidates and their scores, so they know which materials succeeded but not which executable edits caused useful property changes. This makes local refinement difficult when objectives compete and an edit that improves one property may damage another. We propose TRACE, a transition-aware residual control framework that treats evaluated edits as the basic unit of feedback. TRACE records each local refinement as a parent-edit-child transition with observed property deltas, aggregates transition evidence to estimate reusable edit effects, and ranks future edits by their predicted ability to reduce the current candidate's remaining constraint violations while avoiding damage to already satisfied objectives. In a controlled same-backbone comparison, TRACE improves over LLEMA, the state-of-the-art LLM-agent baseline, raising macro-average hit rate from 18.13\% to 25.96\%.
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。