用软约束生成可合成分子，准确率超95%
Synthesizable Molecular Generation via Soft-constrained GFlowNets with Rich Chemical Priors
- 通过软正则化序列型生成流网络，结合化学先验引导分子生成
- 在多种任务中生成可合成分子比例≥95%，且奖励更高
- 适合药物分子设计与自动化合成路径规划的研究者
生成模型在实验性药物发现中的应用受限于从头设计可实际合成分子的难度。以往工作利用生成流网络（GFlowNets）通过基于预定义反应模板和片段的状态与动作空间设计，施加硬性可合成性约束。尽管前景广阔，但该方法灵活性与可扩展性不足。为此，我们提出S3-GFN，通过简单地对基于序列的GFlowNets进行软正则化，生成可合成的SMILES分子。该方法利用大规模SMILES语料库学习到的丰富分子先验，引导生成过程向高奖励、可合成的化学空间推进。模型通过基于可合成与不可合成样本独立缓冲区的离策略回放训练，引入对比学习信号来施加约束。实验表明，S3-GFN在多样任务中能生成≥95%可合成分子，并获得更高奖励。
原文摘要 · Abstract (English)
The application of generative models for experimental drug discovery campaigns is severely limited by the difficulty of designing molecules de novo that can be synthesized in practice. Previous works have leveraged Generative Flow Networks (GFlowNets) to impose hard synthesizability constraints through the design of state and action spaces based on predefined reaction templates and building blocks. Despite the promising prospects of this approach, it currently lacks flexibility and scalability. As an alternative, we propose S3-GFN, which generates synthesizable SMILES molecules via simple soft regularization of a sequence-based GFlowNet. Our approach leverages rich molecular priors learned from large-scale SMILES corpora to steer molecular generation towards high-reward, synthesizable chemical spaces. The model induces constraints through off-policy replay training with a contrastive learning signal based on separate buffers of synthesizable and unsynthesizable samples. Our experiments show that S3-GFN learns to generate synthesizable molecules ($\geq 95\%$) with higher rewards in diverse tasks.
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。