{"ID":2922126,"CreatedAt":"2026-06-02T02:42:49.606572591Z","UpdatedAt":"2026-06-02T15:31:00.366878821Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2606.00776","arxiv_id":"2606.00776","title":"Latent Diffusion Pretraining for Crystal Property Prediction","abstract":"Fast and accurate prediction of crystal properties is a central challenge in new materials design. Graph neural networks and Transformer-based models have emerged as powerful tools for this task due to their ability to encode the local structural environment of atoms within a crystal. However, these models are data-hungry, and in practice, labeled data for crystal properties are scarce. Pretraining-finetuning strategies, particularly those based on diffusion models, have shown promise in addressing these limitations. In this work, we introduce a novel latent diffusion based pretraining framework, CrysLDNet, designed to mitigate data scarcity. Our approach integrates a Variational Autoencoder (VAE) with a diffusion model during the pretraining stage. The VAE encoder maps 3D crystal structures into a smooth latent space within which the diffusion process is applied. This latent diffusion pretraining enables the graph encoder to effectively capture structural and chemical semantics from large-scale unlabeled data, which can then be finetuned for specific property prediction tasks. Comprehensive experiments on popular DFT datasets for property prediction reveal that CrysLDNet significantly outperforms both training-from-scratch and pretrained baselines, with improvements of 4.26% and 4.90% on the JARVIS and MP datasets, respectively. Additionally, the learned representations remain robust in sparse-data conditions and are expressive enough to correct DFT errors when finetuned with limited experimental data. Code is available at: https://github.com/shrimonmuke0202/CrysLDNet.git.","short_abstract":"Fast and accurate prediction of crystal properties is a central challenge in new materials design. Graph neural networks and Transformer-based models have emerged as powerful tools for this task due to their ability to encode the local structural environment of atoms within a crystal. However, these models are data-hun...","url_abs":"https://arxiv.org/abs/2606.00776","url_pdf":"https://arxiv.org/pdf/2606.00776v1","authors":"[\"Shrimon Mukherjee\",\"Kishalay Das\",\"Partha Basuchowdhuri\",\"Pawan Goyal\",\"Niloy Ganguly\"]","published":"2026-05-30T15:44:36Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[\"Graph Neural Network\",\"Diffusion Model\",\"Transformer\",\"Variational Autoencoder\"]","has_code":false,"code_links":[{"ID":612642,"CreatedAt":"2026-06-02T02:42:49.606572591Z","UpdatedAt":"2026-06-02T02:42:49.606572591Z","DeletedAt":null,"paper_id":2922126,"paper_url":"https://arxiv.org/abs/2606.00776","paper_title":"Latent Diffusion Pretraining for Crystal Property Prediction","repo_url":"https://github.com/shrimonmuke0202/CrysLDNet.git","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}