Research map / Graph learning / Graph self-supervised learning
Beyond Structure: Invariant Crystal Property Prediction with Pseudo-Particle Ray Diffraction
Bin Cao, Yang Liu, Longhan Zhang, Yifan Wu, Zhixun Li, Yuyu Luo, Hong Cheng, Yang Ren, Tongyi ZHANG
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add the reciprocal-space diffraction module (ablation) | PRDNet · Materials Project | band gap MAE (eV): 0.361 -> 0.151 | Improved |
| add the reciprocal-space diffraction module (ablation) | PRDNet · Materials Project | formation energy MAE (eV/atom): 0.041 -> 0.028 | Improved |
| add the reciprocal-space diffraction module (ablation) | PRDNet · JARVIS-DFT | bulk modulus MAE (log GPa): 0.287 -> 0.064 | Improved |
| add the reciprocal-space diffraction module (ablation) | PRDNet · Matbench | formation energy MAE (eV/atom): 0.042 -> 0.019 | Improved |
| use multi-head instead of single-head attention (ablation) | PRDNet · Materials Project | band gap MAE (eV): 0.318 -> 0.151 | Improved |
| add residual connections (ablation) | PRDNet · Materials Project | metal/non-metal accuracy (%): 82.7 -> 93.3 | Improved |
| add edge features to attention (ablation) | PRDNet · Materials Project | metal/non-metal accuracy (%): 80.1 -> 93.3 | Improved |
| add edge features to attention (ablation) | PRDNet · JARVIS-DFT | band gap MAE (eV): 0.412 -> 0.140 | Improved |
| use learned pseudo-particle form factors instead of tabulated X-ray form factors (ablation) | PRDNet · Materials Project | shear modulus MAE (log GPa): 0.322 -> 0.108 | Improved |
| use learned pseudo-particle form factors instead of tabulated X-ray form factors (ablation) | PRDNet · Materials Project | metal/non-metal accuracy (%): 78.9 -> 93.3 | Improved |
| attach the diffraction module to baseline CGCNN | CGCNN · Materials Project | band gap MAE (eV): 0.262 -> 0.207 | Improved |
| attach the diffraction module to baseline GATGNN | GATGNN · Materials Project | shear modulus MAE (log GPa): 0.111 -> 0.169 | Hurt |
Closest papers by meaning
- All that structure matches does not glitter (NeurIPS 2025 · similarity 0.88)
- Local-Global Associative Frames for Symmetry-Preserving Crystal Structure Modeling (NeurIPS 2025 · similarity 0.87)
- WyckoffDiff -- A Generative Diffusion Model for Crystal Symmetry (ICML 2025 · similarity 0.87)
- A Diffusion-Based Pre-training Framework for Crystal Property Prediction (AAAI 2024 · similarity 0.87)
- Complete and Efficient Graph Transformers for Crystal Material Property Prediction (ICLR 2024 · similarity 0.87)
- Learning the Electronic Hamiltonian of Large Atomic Structures (ICML 2025 · similarity 0.86)
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.
