用块扩散模型生成药物分子，兼顾结构合理性和设计效率。
From Tokens to Blocks: A Block-Diffusion Perspective on Molecular Generation
- 提出软片段表示与块扩散架构，突破传统文本生成限制。
- 生成分子100%合法，亲和力提升9.7%，多样性增2-3倍。
- 适合新药设计、生成模型研究者快速部署使用。
药物发现可视为在巨大化学空间中的组合搜索，推动了深度生成模型在全新分子设计中的应用。基于GPT的分子语言模型（MLM）通过大规模数据学习化学语法规则，表现出色。但现有模型存在两大局限：将分子建模为逐词预测时未能充分捕捉其图结构特性；且通常缺乏面向目标的生成机制。为此，本文提出SoftMol统一框架，协同设计分子表示、模型架构与搜索策略，实现目标导向分子生成。SoftMol引入无规则软片段表示，支持原生扩散建模，并开发首个块扩散分子语言模型SoftBD，结合局部双向扩散与自回归生成，在分子结构约束下完成生成。为提升药物相似性与合成可行性，SoftBD在精心筛选的数据集ZINC-Curated上训练。SoftMol进一步集成门控蒙特卡洛树搜索，实现目标导向片段组装。实验表明，相比当前最先进模型，SoftMol实现100%化学合法性，结合亲和力提升9.7%，分子多样性提高2-3倍，推理效率提升6.6倍。代码已开源。
原文摘要 · Abstract (English)
Drug discovery can be viewed as a combinatorial search over an immense chemical space, motivating the development of deep generative models for de novo molecular design. Among these, GPT-based molecular language models (MLM) have shown strong molecular design performance by learning chemical syntax and semantics from large-scale data. However, existing MLMs face two fundamental limitations: they inadequately capture the graph-structured nature of molecules when formulated as next-token prediction problems, and they typically lack explicit mechanisms for target-aware generation. Here, we propose SoftMol, a unified framework that co-designs molecular representation, model architecture, and search strategy for target-aware molecular generation. SoftMol introduces soft fragments, a rule-free block representation of SMILES that enables diffusion-native modeling, and develops SoftBD, the first block-diffusion molecular language model that combines local bidirectional diffusion with autoregressive generation under molecular structural constraints. To favor generated molecules with high drug-likeness and synthetic accessibility, SoftBD is trained on a carefully curated dataset named ZINC-Curated. SoftMol further integrates a gated Monte Carlo tree search to assemble fragments in a target-aware manner. Experimental results show that, compared with current state-of-the-art models, SoftMol achieves 100% chemical validity, improves binding affinity by 9.7%, yields a 2-3x increase in molecular diversity, and delivers a 6.6x speedup in inference efficiency. Code is available at https://github.com/szu-aicourse/softmol
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。