PaddleMaterials is a data-mechanism dual-driven, foundation model development and deployment, end to end toolkit based on PaddlePaddle deep learning framework for materials science.

[ICLR 2026] Advancing Universal Deep Learning for Electronic-Structure Hamiltonian Prediction of Materials

This repository is a collection of my thoughts, summaries, and reflections on literature, websites, and tutorials that explore the intersection of materials science and artificial intelligence.

Public repository of ACL 2026 system demo paper, PhyVer: Physics-Grounded Material Claim Verification with Multi-Fidelity Physical Evidence.