Solvent:用于多步逆合成的长程推理模型
Solvent: Long-Horizon Reasoning for Multi-Step Retrosynthesis
AI总结:
本研究推出8B参数LLM Solvent,通过两轮后训练课程整合迭代合成规划等功能,经RL获2.1倍性能提升,在四个基准测试中得分61.86%,超越同类模型并可媲美前沿万亿参数LLM,实现多步逆合成的长程推理。
摘要:
多步逆合成需要通过连续反应步骤将目标分子递归转化为可购买的构建模块,近期研究开始借助大语言模型(LLM)的推理能力解决该任务。最新进展中,LLM推理已进入长程推理阶段:模型在相互依赖的步骤中追求目标、与环境交互并适应反馈,这一转变显著提升了LLM的能力,如AI驱动数学研究的突破所示。在该前沿领域,我们推出Solvent,一款8B参数的LLM,成功将长程推理扩展至多步逆合成。通过两轮后训练课程,Solvent学会将迭代合成规划、外部可用性验证与自适应路线修订相结合。实验显示,经RL带来2.1倍性能提升后,Solvent在四个基准测试中得分61.86%,超过中等规模语言模型及此前基于搜索的逆合成系统,且可与前沿万亿参数LLM相媲美。
英文摘要:
Multi-step retrosynthesis requires recursively transforming a target molecule through successive reaction steps into purchasable building blocks, and recent work has begun to draw on large language model (LLM) reasoning to tackle the task. In its latest evolution, LLM reasoning has entered the long-horizon regime: models pursue objectives across interdependent steps, interact with environments, and adapt to feedback. This shift has markedly advanced LLM capabilities, as illustrated by breakthroughs in AI-driven mathematical research. At this frontier, we introduce Solvent, an 8B LLM that successfully extends long-horizon reasoning to multi-step retrosynthesis. Through a two-round post-training curriculum, Solvent learns to interleave iterative synthesis planning, external availability verification, and adaptive route revision. Experiments show that, following a $2.1\times$ performance boost from RL, Solvent scores 61.86% across four benchmarks, surpassing mid-scale language models and prior search-based retrosynthesis systems, and rivaling frontier trillion-parameter LLMs.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。