arXiv cs.AI· Daniel T. Levy, Vineet Jain, Tara Akhound-Sadegh, S\'ekou-Oumar Kaba, Siamak Ravanbakhsh·· 4 小时前AI 评分41
扩散树采样(DTS)实现晶体生成模型的推理时适配
Diffusion Tree Sampling for Inference Time Adaptation of Crystal Generative Models
AI 导读
一篇 NeurIPS 2026 论文提出 Diffusion Tree Sampling(DTS),在去噪轨迹上运行蒙特卡洛树搜索,无需微调即可在推理时适配扩散式晶体生成模型。
正文 · 原文
Abstract:Diffusion-based generative models have become strong priors over plausible crystal structures, but turning that prior into promising candidates requires steering toward several conflicting properties at once. Fine-tuning is expensive and produces a new model for every new objective, while differentiable guidance is not applicable to many real-world tasks. We instead adapt these generative models at inference time using Diffusion Tree Sampling (DTS), a Monte Carlo tree search procedure over the denoising trajectory. We make two technical contributions on top of DTS that are particularly relevant for AI-for-science applications. First, we show how to use existing property-conditional generative models (e.g., models fine-tuned for classifier-free guidance) as guided proposals inside DTS, without distorting the target distribution. Second, we extend DTS to multi-objective sampling, so that a single search produces samples spanning an entire Pareto front. We instantiate both adaptations on de novo crystal generation: improving the stability of crystals generated by different models, using pretrained conditional models to generate crystals belonging to rare space groups or with high bulk modulus, and generating crystals along a Pareto front of high dielectric constant and high bandgap materials. Across these tasks, DTS-based inference-time adaptation successfully improves model outputs without modifying the models themselves.
| Comments: | 24 pages, 7 figures. Accepted to NeurIPS 2026 |
| Subjects: | Machine Learning (cs.LG); Materials Science (cond-mat.mtrl-sci); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.17462 [cs.LG] |
| (or arXiv:2611.17462v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2611.17462 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Daniel Levy [view email]
[v1]
Thu, 12 Nov 2026 19:40:38 UTC (1,978 KB)
来源:arXiv cs.AI · arxiv.org