arXiv cs.AI· Yogesh Verma, Dani Korpela, Harri L\"ahdesm\"aki, Vikas Garg·· 1 天前AI 评分31
面向生物序列设计的灵活流匹配方法
Flexible Flows for Biological Sequence Design
AI 导读
研究者提出一种结构化耦合的离散流匹配（DFM）方法，通过编码序列元素间的领域偏好将源分布偏向合理区域，且不改变流目标或训练流程。该方法引入基于编辑操作的隐变量速率参数化实现可变长度生成，并用隐空间无分类器引导与 Dirichlet 先验温度缩放实现测试时控制。在密度估计、无条件与条件 DNA 序列生成及多肽序列生成等任务上达到 SOTA，代码已开源。
正文 · 原文
Abstract:Designing functional biological sequences requires navigating vast discrete spaces under strict evolutionary and biophysical constraints. Discrete Flow Matching (DFM) offers a generative framework over such spaces, but existing approaches rely on biologically uninformative couplings and offer limited flexibility for variable-length sequence generation and fine-grained control. We propose a structured coupling that encodes domain-specific preferences among sequence elements, biasing the source distribution toward plausible regions without modifying the flow objective or training procedure. Building on this, we introduce a latent edit-based rate parameterization that models variable-length generation via edit operations conditioned on a shared global latent, akin to a latent variable model, while remaining tractable. We further introduce a latent classifier-free guidance mechanism that steers generation coherently in continuous latent space, along with Dirichlet-prior temperature scaling for test-time control over edit operations. Our method achieves state-of-the-art performance across diverse biological sequence tasks, including density estimation, unconditional and conditional DNA sequence generation, and peptide sequence generation. Our code is available at this https URL.
| Comments: | Accepted at NeurIPS 2026 |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Emerging Technologies (cs.ET); Quantitative Methods (q-bio.QM) |
| Cite as: | arXiv:2606.10543 [cs.LG] |
| (or arXiv:2606.10543v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2606.10543 arXiv-issued DOI via DataCite |
Submission history
From: Yogesh Verma [view email]
[v1]
Tue, 9 Jun 2026 08:11:14 UTC (3,884 KB)
[v2]
Wed, 11 Nov 2026 10:58:03 UTC (3,884 KB)
来源：arXiv cs.AI · arxiv.org