Newest papers

MFG: Multi-turn Feedback Guided Molecular Optimization

Trieu Nguyen, Frazier N. Baker, Ruoxi Gao, Daniel Adu‐Ampratwum, Xia Ning · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.18851 · cited by 0 works (OpenAlex) · open-access PDF

French

L'optimisation moléculaire multi-objectif est un processus critique dans la découverte de médicaments qui vise à modifier une molécule donnée pour améliorer simultanément plusieurs propriétés tout en conservant une structure globale similaire. Ce processus nécessite souvent des modifications itératives guidées par des commentaires computationnels ou expérimentaux. Les méthodes récentes basées sur le grand modèle de langue (LLM) ont commencé à explorer cette approche itérative; cependant, ces méthodes ne sont pas explicitement formées pour tirer parti de tels commentaires, tandis que les approches existantes se concentrent principalement sur l'optimisation moléculaire à tour unique. Pour résoudre ce fossé, nous introduisons l'optimisation moléculaire Multi-turn Feedback Guided (MFG

Machine translation (Meta M2M100); the English original follows.

Abstract

Multi-objective molecular optimization is a critical process in drug discovery that aims to modify a given molecule to simultaneously improve multiple properties while maintaining a similar overall structure. This process often requires iterative modifications guided by computational or experimental feedback. Recent large language model (LLM)-based methods have begun to explore this iterative approach; however, these methods are not explicitly trained to leverage such feedback, while existing approaches primarily focus on single-turn molecular optimization. To address this gap, we introduce Multi-turn Feedback Guided (MFG) molecular optimization, a reinforcement learning framework that trains an LLM to progressively optimize molecules using feedback from previous turns. MFG introduces a feedback vector capturing both property improvements and source molecule similarity, alongside a feedback-aligned reward function that promotes progressive improvement across turns. We apply MFG over $\text{GeLLM}^{3}\text{O}$, a state-of-the-art LLM model for multi-objective molecular optimization, evaluate $\text{GeLLM}^{3}\text{O-MFG}$ on ten multi-objective molecular optimization tasks. The results demonstrate that $\text{GeLLM}^{3}\text{O-MFG}$ achieves the highest average similarity-constrained success rate ($81.50%$), outperforming the base model $\text{GeLLM}^{3}\text{O}$ ($70.81%$).

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Leverage (statistics), Process (computing), Reinforcement learning, Iterative and incremental development, Property (philosophy)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange