Rachel：一种通用语言模型指导并修正逆合成路线
Rachel: A general-purpose language model directs and revises retrosynthetic routes
浏览论文内容
中文总结 AI 辅助
本文提出Rachel环境，验证通用LLM GPT-5.5无需搜索策略即可自主规划并修正逆合成路线，在PaRoutes120和RF25上实现高严格闭合率，优于多数比较方法。
中文摘要 AI 辅助
逆合成规划通过重塑剩余化学问题的决策来推进：一个局部合理的断键可能留下前体，其化学选择性约束使路线的其余部分复杂化。现有规划器通常将模型提议引导至搜索或模板程序，这留下了一个问题：通用大型语言模型（LLM）本身能否维持并修正路线策略。我们开发了Rachel，一个状态化环境，它执行并检查LLM指导的化学操作，但不规定搜索策略或停止规则。在没有提供参考路线或路线级解决方案的情况下，GPT-5.5在PaRoutes120目标中的111个以及独立RF25困难目标队列中的25个目标中的24个实现了严格闭合。RF25主要取自GPT-5.5报告的知识截止日期之后发表的研究。闭合要求完整路线以及规划后每个末端前体的独立来源解析。在共享的PaRoutes子集上，前向模型支持超过了大多数比较方法，Rachel从两个方法盲法的LLM评估者那里获得了最高的平均总体路线得分。记录的轨迹显示了持续的模型提议化学操作，修订后的策略被带入后续步骤。用固定策略替换LLM路线决策将严格闭合降低到6-15/120，尽管局部化学执行仍在继续；限制规划支持也降低了RF25中的闭合。在Rachel内部，一个通用LLM协调了连续的化学选择，并在早期决策重塑剩余问题时修正其策略。
英文摘要
Retrosynthetic planning advances through decisions that reshape the remaining chemical problem: a locally plausible disconnection can leave precursors whose chemoselectivity constraints complicate the rest of the route. Existing planners often channel model proposals through search or template procedures, leaving open whether a general-purpose large language model (LLM) can itself sustain and revise route strategy. We developed Rachel, a stateful environment that executes and checks LLM-directed chemistry but prescribes neither a search policy nor a stopping rule. Without supplied reference routes or route-level solutions, GPT-5.5 achieved strict closure for 111 of 120 PaRoutes120 targets and 24 of 25 targets in the separate RF25 difficult-target cohort. RF25 was drawn largely from studies published after GPT-5.5's reported knowledge cutoff. Closure required complete routes and independent source resolution of every terminal precursor after planning. On a shared PaRoutes subset, forward-model support exceeded that of most comparator methods, and Rachel received the highest mean overall route score from both method-blinded LLM evaluators. Recorded trajectories showed continued model-proposed chemistry, with revised strategies carried into subsequent steps. Replacing LLM route decisions with fixed policies reduced strict closure to 6-15/120 despite continued local chemical execution; restricting planning support also reduced closure in RF25. Within Rachel, a general-purpose LLM coordinated successive chemical choices and revised its strategy as earlier decisions reshaped the remaining problems.
发表机构
- Hangzhou Institute of Medicine, Chinese Academy of Sciences（中国科学院杭州医学研究所）
- Beijing University of Chemical Technology（北京化工大学）
- Zhenzhida Biotechnology (Shanghai) Co., Ltd.（臻智达生物科技（上海）有限公司）
- Renji Hospital, School of Medicine, Shanghai Jiao Tong University（上海交通大学医学院附属仁济医院）
机构由 AI 辅助整理，请以论文原文为准。
