通过提示工程微调大模型，实现定向药物分子生成。
SmileyLlama: Modifying Large Language Models for Directed Chemical Space Exploration
- 用指令微调让大模型直接扮演化学语言模型。
- 生成分子符合用户指定属性，且结合强化学习优化三维构象与亲和力。
- 适用于药物研发，框架可扩展至材料与生物领域。
我们展示通过监督微调（SFT）设计的提示，可将大语言模型（LLM）转化为用于探索药物分子化学空间的SmileyLlama。在生成有效且新颖的类药物分子方面，我们对SmileyLlama与预训练LLM及从零训练的化学语言模型（CLM）进行了基准测试，并使用直接偏好优化（DPO）提升其对提示的遵循度，同时作为iMiner强化学习框架的一部分，预测具有优化3D构象和高结合亲和力的分子。通过训练大模型以化学语言模型身份直接生成分子，同时保留其自然语言能力，我们实现了根据用户需求可靠生成特定属性分子，而非仅作为具备化学知识的聊天机器人或虚拟助手。尽管SmileyLlama聚焦于药物发现，但该SFT/DPO/LLM框架可推广至其他化学、生物和材料应用。
原文摘要 · Abstract (English)
We show that large language model (LLMs) can be transformed via supervised fine-tuning (SFT) of engineered prompts into SmileyLlama for exploring the chemical space of drug molecules. We benchmark SmileyLlama against pre-trained LLMs and chemical language models (CLM) trained from scratch for generating valid and novel drug-like molecules, and use direct preference optimization (DPO) to both improve SmileyLlama's adherence to a prompt and as part of the iMiner reinforcement learning framework to predict molecules with optimized 3D conformations and high binding affinity to drug targets. By training an LLM to speak directly as a CLM, while retaining most of its natural language capabilities, we show that we can reliably generate molecules with user-specified properties rather than acting only as a chatbot with knowledge of chemistry or as a virtual assistant. While SmileyLlama is geared toward drug discovery, the SFT/DPO/LLM framework can be extended to other chemical, biological, and materials applications.
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。