AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Geometric Latent Diffusion Models for 3D Molecule Generation

Minkai Xu, Alexander S Powers, Ron O. Dror, Stefano Ermon, Jure Leskovec

ICML 2023 · Image generation › Autoregressive and latent generation · approach: Equivariant diffusion for 3D molecules · 31 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use equivariant point-structured latent features rather than invariant-only latents (ablation)GeoLDM · QM9validity (%): 83.6 -> 93.8Improved
Use equivariant point-structured latent features rather than invariant-only latents (ablation)GeoLDM · GEOM-DRUGvalidity (%): 97.2 -> 99.3Improved
Add random-rotation augmentation while using invariant-only latent modeling (ablation)GraphLDM · QM9validity (%): 83.6 -> 90.5Improved
Add random-rotation augmentation while using invariant-only latent modeling (ablation)GraphLDM · GEOM-DRUGvalidity (%): 97.2 -> 98.0Improved
Use ES-reg early stopping instead of KL-reg (ablation)GeoLDM (k=1) · QM9validity (%): 83.7 -> 93.8Improved
Reduce latent invariant dimension from k=16 to k=1 (ablation)GeoLDM with ES-reg · QM9validity (%): 92.4 -> 93.8Improved
Reduce latent invariant dimension from k=8 to k=1 (ablation)GeoLDM with ES-reg · QM9validity (%): 92.1 -> 93.8Improved
Reduce latent invariant dimension from k=4 to k=1 (ablation)GeoLDM with ES-reg · QM9validity (%): 92.6 -> 93.8Improved
Model generation in equivariant latent space instead of atomic feature spaceGeoLDM vs EDM · QM9 conditional generationMAE (Bohr^3): 2.76 -> 2.37Improved
Model generation in equivariant latent space instead of atomic feature spaceGeoLDM vs EDM · QM9 conditional generationMAE (meV), Δε: 655 -> 587Improved
Model generation in equivariant latent space instead of atomic feature spaceGeoLDM vs EDM · QM9 conditional generationMAE (meV), εHOMO: 356 -> 340Improved
Model generation in equivariant latent space instead of atomic feature spaceGeoLDM vs EDM · QM9 conditional generationMAE (meV), εLUMO: 584 -> 522Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy