technique

E(3)-equivariant GNN

techniqueactivee-3-equivariant-gnn-4d18d967·1 events·first seen 27d ago

Aliases: E(3)-equivariant GNN

Co-occurring entities

EvoStruct protein language models complementarity-determining region (CDR) design CHIMERA-Bench R-Drop consistency regularization cross-attention adapter

More like this (12)

E(3) equivariant graph neural network Graph Neural Networks Generative Adversarial Networks Graph Neural Network leveraging Higher-order Class Label Connectivity for Heterophilous Graphs Graph Neural Network Encoder Graph Structure Learning neural network image classifiers EAGLE-3 NNCF (Neural Network Compression Framework)Recurrent Neural Network Spiking Neural Networks Graph Transformer

Recent events (1)

5arXiv · cs.LG·27d ago·source ↗

EvoStruct: Bridging Evolutionary and Structural Priors for Antibody CDR Design via Protein Language Model Adaptation

EvoStruct addresses vocabulary collapse in GNN-based antibody CDR design by combining a frozen protein language model with an E(3)-equivariant GNN through a cross-attention adapter. The method introduces progressive PLM unfreezing and R-Drop consistency regularization to recover functionally important amino acid diversity. On CHIMERA-Bench, EvoStruct improves sequence recovery by 16%, reduces perplexity by 43%, and achieves 2.3x greater amino acid diversity compared to the best GNN baselines.

Evaluation and Benchmarking Multimodal Progress EvoStruct protein language models E(3)-equivariant GNN +4 more