为分子生成扩散模型提供每样本不确定性评估，判断生成质量。
Uncertainty Estimation for Molecular Diffusion Models
- 基于去噪网络的拉普拉斯近似，追踪生成过程中的噪声预测变化。
- 不确定性分数与分子质量指标负相关，分数越低质量越差。
- 可用于筛选生成样本，提升模型测试时性能，适合分子设计应用。
扩散模型在三维分子生成中广泛应用，但缺乏对生成分子质量低下的可靠预警信号。本文提出一种针对预训练分子扩散模型的后验不确定性估计方法。基于去噪网络的拉普拉斯近似，测量生成轨迹中噪声预测的变异性，得到每样本的不确定性得分。实验表明，该得分能有效反映生成样本质量，与现有样本级质量指标呈负相关。进一步研究显示，利用该不确定性分数可对生成样本进行过滤，通过测试时缩放提升模型性能。
原文摘要 · Abstract (English)
Diffusion models have seen wide adoption for 3D molecular generation, yet they offer no principled signal of when a generated molecule is likely to be of low quality. We propose a post-hoc method for estimating per-sample uncertainty in pretrained molecular diffusion models. Building on a Laplace approximation of the denoising network, we measure the variability of the noise prediction across the generation trajectory. Empirically, we show that the resulting uncertainty score is informative of sample quality, exhibiting a negative correlation with established sample-level quality metrics. We further study how the proposed uncertainty score can be used to filter generated samples, improving model performance via test-time scaling.
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。