研究人员推出BiMol-Diff，一个新颖的扩散框架，旨在连接分子结构和自然语言以实现可控设计。该方法利用了感知token的噪声调度，根据token恢复难度调整损坏级别，以更好地保留关键子结构。BiMol-Diff在分子重建方面表现出改进，在基准数据集上实现了15.4%的精确匹配相对提升，并在分子描述任务中也展现出强劲性能。 AI
影响 引入了一种新的分子结构-语言建模方法，可能改进AI驱动的药物发现和设计。
排序理由 介绍用于分子生成和描述的新框架的学术论文。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
arXiv:2604.24089v1 Announce Type: new Abstract: Bridging molecular structures and natural language is essential for controllable design. Autoregressive models struggle with long-range dependencies, while standard diffusion processes apply uniform corruption across positions, whic…
Bridging molecular structures and natural language is essential for controllable design. Autoregressive models struggle with long-range dependencies, while standard diffusion processes apply uniform corruption across positions, which can distort structurally informative tokens. W…
Bridging molecular structures and natural language is essential for controllable design. Autoregressive models struggle with long-range dependencies, while standard diffusion processes apply uniform corruption across positions, which can distort structurally informative tokens. W…