CrysLDNet combines VAE and latent diffusion pretraining on unlabeled crystals to improve graph encoder performance on property prediction by about 4-5% on JARVIS and MP datasets.
Connectivity optimized nested line graph networks for crystal structures
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
An autonomous LLM coding agent built the top-performing crystal graph network on the MatBench band-gap benchmark by implementing known methods, outperforming 17 expert models without pretraining.
citing papers explorer
-
Latent Diffusion Pretraining for Crystal Property Prediction
CrysLDNet combines VAE and latent diffusion pretraining on unlabeled crystals to improve graph encoder performance on property prediction by about 4-5% on JARVIS and MP datasets.
-
Optimizing Expert-Designed Crystal Graph Networks for Band-Gap Prediction with an Autonomous LLM Research Loop
An autonomous LLM coding agent built the top-performing crystal graph network on the MatBench band-gap benchmark by implementing known methods, outperforming 17 expert models without pretraining.