FlashCart: Fast Cartesian Tensor Products for Equivariant Interatomic Potentials
cs.LG
Submitted: 2026-10-05
Updated: 2026-10-05
Sources
- Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphs
- DPA4: Pushing the Accuracy-Cost Frontier of Interatomic Potentials with EMFA SO(2) Convolution
- An Efficient Sparse Kernel Generator for O(3)-Equivariant Deep Networks
- Sobek: Streaming Equivariant Tensor Product Convolutions
- Cormorant: Covariant Molecular Neural Networks
- Orb: A Fast, Scalable Neural Network Potential
- A Cartesian-3j Framework for Machine Learning Interatomic Potentials
- EquiformerV3: Scaling Efficient, Expressive, and General SE(3)-Equivariant Graph Attention Transformers
- 3D Steerable CNNs: Learning Rotationally Equivariant Features in Volumetric Data
- Directional Message Passing for Molecular Graphs
- MACE-POLAR-1: A Polarisable Electrostatic Foundation Model for Molecular Chemistry
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks