AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Latent Diffusion Pretraining for Crystal Property Prediction

Shrimon Mukherjee, KISHALAY DAS, Partha Basuchowdhuri, Pawan Goyal, Niloy Ganguly

ICML 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Machine-learned interatomic potentials · 0 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
add latent diffusion pretraining on top of PDDFormer encoderPDDFormer · JARVIS-DFT Bulk ModulusMAE: 9.546 -> 8.817Improved
add latent diffusion pretraining on top of Matformer encoderMatformer · JARVIS-DFT EhullMAE: 0.064 -> 0.045Improved
do latent-space diffusion instead of feature-space diffusion (DPF)crystal pretraining · JARVIS-DFT Bandgap (MBJ)MAE: 0.311 -> 0.238Improved
add LDM stage after VAE pretraining (ablation)CrysLDNet pretraining · JARVIS-DFT Bulk ModulusMAE: 10.61 -> 8.817Improved
add VAE pretraining before latent diffusion (ablation)CrysLDNet pretraining · JARVIS-DFT Bandgap (MBJ)MAE: 0.302 -> 0.238Improved
LDM-only pretraining versus VAE-only pretraining (ablation)CrysLDNet pretraining · JARVIS-DFT (4 properties)MAE: Bulk 10.61 (VAE) vs 10.37 (LDM); MBJ 0.284 (VAE) vs 0.302 (LDM)Mixed
reconstruct all of atom types, coordinates and lattice in the VAE instead of a subset (ablation)CrysLDNet VAE · JARVIS-DFT SLMEMAE: 4.678 -> 4.120Improved
train LDM longer to refine VAE latents (5 to 50 epochs) (ablation)CrysLDNet (Matformer) · JARVIS-DFT Bulk ModulusMAE: 10.35 -> 9.818Improved
latent diffusion refinement raises mutual information of latents (ablation)VAE-only encoder · GNoME pretraining set, MI with atom typesmutual information: 3.0906 -> 4.5465Improved
pretrain with CrysLDNet when finetuning data is scarce (20-40%)Matformer and PDDFormer · JARVIS-DFT limited dataMAE: average improvement 12.83% and 22.49% at 40% dataImproved
pretrain with CrysLDNet when data is plentiful (large property sets)PDDFormer · JARVIS-DFT large-data propertiesMAE: average gain 3.0% (large data) vs 5.27% (small data)Improved
pretrain with CrysLDNet then finetune on DFT, zero-shot on experimentCrysGNN, CrysDiff, DPF · OQMD-EXP formation energyMAE: 0.211 (CrysDiff) -> 0.205Improved

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