arXiv:2511.09568v2 Announce Type: replace-cross 
Abstract: Diffusion models show promise for 3D molecular generation, but face a fundamental trade-off between sampling efficiency and conformational accuracy. While flow-based models are fast, they often produce geometrically inaccurate structures, as they have difficulty capturing the multimodal distributions of molecular conformations. In contrast, denoising diffusion models are more accurate but suffer from slow sampling, a limitation attributed to sub-optimal integration between diffusion dynamics and SE(3)-equivariant architectures. To address this, we propose VEDA, a unified SE(3)-equivariant framework that combines variance-exploding diffusion with annealing to efficiently generate conformationally accurate 3D molecular structures. Specifically, our key technical contributions include: (1) a VE schedule that enables noise injection functionally analogous to simulated annealing, improving 3D accuracy and reducing relaxation energy; (2) a novel preconditioning scheme that reconciles the coordinate-predicting nature of SE(3)-equivariant networks with a residual-based diffusion objective, and (3) a new arcsin-based scheduler that concentrates sampling in critical intervals of the logarithmic signal-to-noise ratio. On the QM9 and GEOM-DRUGS datasets, VEDA matches the sampling efficiency of flow-based models, achieving state-of-the-art valency stability and validity with only 100 sampling steps. More importantly, VEDA's generated structures are remarkably stable, as measured by their relaxation energy during GFN2-xTB optimization. The median energy change is only 1.72 kcal/mol, significantly lower than the 32.3 kcal/mol from its architectural baseline, SemlaFlow. Our framework demonstrates that principled integration of VE diffusion with SE(3)-equivariant architectures can achieve both high chemical accuracy and computational efficiency.

تم تقديم VEDA كإطار جديد لتوليد الجزيئات ثلاثية الأبعاد، باستخدام انتشار متفجر للمتغيرات مع التلدين لتحسين دقة الهياكل الجزيئية مع تحسين كفاءة العينة. تتناول هذه الطريقة قيود النماذج الحالية، التي تكافح إما مع السرعة أو الدقة في توليد التشكيلات الجزيئية.

VEDA se ha presentado como un nuevo marco para la generación de moléculas 3D, utilizando la difusión de varianza explosiva combinada con recocido para mejorar la precisión de las estructuras moleculares mientras se mejora la eficiencia de muestreo. Este enfoque aborda las limitaciones de los modelos existentes, que luchan con la velocidad o la precisión en la generación de conformaciones moleculares.

VEDA a été introduit comme un nouveau cadre pour la génération de molécules 3D, utilisant la diffusion à variance explosive combinée à un recuit pour améliorer l'exactitude des structures moléculaires tout en améliorant l'efficacité d'échantillonnage. Cette approche répond aux limitations des modèles existants, qui peinent soit avec la vitesse, soit avec l'exactitude dans la génération de conformations moléculaires.

VEDA has been introduced as a new framework for 3D molecular generation, utilizing variance-exploding diffusion combined with annealing to enhance the accuracy of molecular structures while improving sampling efficiency. This approach addresses the limitations of existing models, which struggle with either speed or accuracy in generating molecular conformations.

VEDA: 3D Molecular Generation via Variance-Exploding Diffusion with Annealing

Was this article worth reading? Share it

PlantFCE Model Builder

LucidQuery AI

VECTARY

Metaflow AI

Dynamiq

raz-vape

Ready to build your own newsroom?