Tag
Caltech researchers introduce eSpinor, an SU(2)-equivariant graph neural network that propagates spinor features to predict spin-orbit-coupled magnetic properties of molecules, improving magnetic anisotropy and excitation-energy predictions for single-molecule magnets and MRI probes while generalizing to unseen ligands.
LEMON is a graph neural network monitor that uses equivariant learning over attention maps to detect LLM misbehavior zero-shot across unseen models, outperforming baselines by up to 14 ROC-AUC points on hallucination and jailbreak monitoring across 16 LLMs and six datasets.
O-kRystal, a periodic E(3)-equivariant graph neural network, learns corrections to electronic states beyond DFT, achieving HSE06 band predictions across nearly 6,000 materials, G0W0 quasiparticle energies in 2D systems, and a 68.7% reduction in band energy MAE on strongly correlated materials.
NEMORA 是一种将 Fast Multipole Method 推广为可学习等变算子的新方法，用于在等变图神经网络中建模长程原子间相互作用，具备线性时间与内存复杂度，可处理数十万原子体系，并在长程基准测试上将力和能量误差降低一个至三个数量级。
The paper proposes Autoregressive Frontier Expansion, a generative framework that grows tree-like branching structures iteratively, using a flow-matching model parameterized by an SO(2)-equivariant GNN to decide bifurcation versus termination at each frontier step. It is evaluated on cortical neuron and botanical tree morphology generation in unconditional, class-conditioned, and morphology-guided settings, closely matching reference distributions.