arXiv：cs.LG· Yang Qiao, Bo Pan, Hao-Wei Pang, Peter Zhiping Zhang, Liying Zhang, Liang Zhao·· 5 小时前AI 评分32
MolWorld：面向可执行分子优化的分子世界模型
MolWorld: Molecule World Models for Actionable Molecular Optimization
AI 导读
研究者提出 MolWorld，一个由分子世界模型引导的框架，将可执行分子优化建模为分子迁移图的迭代扩展，节点为分子、边为匹配分子对（MMP）关系。该框架用连通局部子图作为锚点上下文，由潜在世界模型预测局部结构扩展，再由上下文条件生成器提出候选分子，经 MMP 关系验证后并入图以保持可达性。在性质优化与基于对接的任务上，MolWorld 能发现高性质分子并维持更强的结构连通性。
正文
Abstract:Molecular optimization in drug discovery aims to discover molecules with improved target properties, but practical lead optimization often requires more than high predicted scores. A useful candidate should also be actionable: it should be reachable from known molecules through a sequence of local structural modifications, providing explicit structural references for interpreting property changes within an evolving chemical series. Existing de novo and single-molecule optimization methods do not explicitly model such reachability, especially when both the target molecules and the intermediate molecules connecting them to known compounds are unknown. In this work, we formulate actionable molecular optimization as iterative expansion of a molecule-transfer graph, where nodes are molecules and edges encode matched molecular pair (MMP) relations representing localized structural differences. We propose MolWorld, a molecule world model-guided framework that uses connected local subgraphs as anchor contexts for graph expansion. Given the current graph and a candidate context, a latent molecule world model predicts the resulting local structural expansion. A context-conditioned generator then uses the molecular structures and MMP relations within selected contexts to propose candidates, learning from local graph-completion tasks. Candidates are evaluated, and those connected through verified MMP relations are incorporated into the graph, preserving reachability from the initial set. The updated graph serves as the state for subsequent optimization. Experiments on property optimization and docking-based tasks show that MolWorld discovers high-property molecules while maintaining stronger structural connectivity, supporting actionable and sequential molecular design.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2605.08954 [cs.LG] |
| (or arXiv:2605.08954v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2605.08954 arXiv-issued DOI via DataCite |
Submission history
From: Yang Qiao [view email]
[v1]
Sat, 9 May 2026 13:50:56 UTC (5,762 KB)
[v2]
Thu, 1 Oct 2026 19:29:28 UTC (1,362 KB)
来源：arXiv：cs.LG · arxiv.org