mlsanityData to 7 Oct 2026

GNN-MoE: Context-Aware Patch Routing using GNNs for Parameter-Efficient Domain Generalization

arXiv 2511.04008 ↗First posted 6 Nov 2025cs.CVcited by 0

On the map: Mixture-of-experts → Vision mixture-of-experts

Abstract (arXiv)

Domain generalization (DG) seeks robust Vision Transformer (ViT) performance on unseen domains. Efficiently adapting pretrained ViTs for DG is challenging; standard fine-tuning is costly and can impair generalization. We propose GNN-MoE, enhancing Parameter-Efficient Fine-Tuning (PEFT) for DG with a Mixture-of-Experts (MoE) framework using efficient Kronecker adapters. Instead of token-based routing, a novel Graph Neural Network (GNN) router (GCN, GAT, SAGE) operates on inter-patch graphs to dynamically assign patches to specialized experts. This context-aware GNN routing leverages inter-patch relationships for better adaptation to domain shifts. GNN-MoE achieves state-of-the-art or competitive DG benchmark performance with high parameter efficiency, highlighting the utility of graph-based contextual routing for robust, lightweight DG.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Vision mixture-of-experts, one of 12 directions in Mixture-of-experts. Builds on 18 of its 21 references that are in the map. First posted on arXiv 6 Nov 2025.

Builds on

18 of 21 references.

References · 18

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  3. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  4. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  5. 2017Graph Attention Networkscited by 8,854
  6. 2017Inductive Representation Learning on Large Graphscited by 6,284
  7. 2019Parameter-Efficient Transfer Learning for NLPcited by 4,195
  8. 2017Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layercited by 3,413
  9. 2021Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsitycited by 3,014
  10. 2019Fast Graph Representation Learning with PyTorch Geometriccited by 2,414
  11. 2018Moment Matching for Multi-Source Domain Adaptationcited by 1,327
  12. 2017Deep Hashing Network for Unsupervised Domain Adaptationcited by 1,181
6 more
  1. 2017Deeper, Broader and Artier Domain Generalizationcited by 983
  2. 2021Robust fine-tuning of zero-shot modelscited by 785
  3. 2021Domain Generalization: A Surveycited by 760
  4. 2018Recognition in Terra Incognitacited by 655
  5. 2022Self-Distilled Vision Transformer for Domain Generalizationcited by 35
  6. 2023Domain Generalization Using Large Pretrained Models with Mixture-of-Adapterscited by 11

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Generalizing GNNs with Tokenized Mixture of Experts
  2. 2022Sparse Mixture-of-Experts are Domain Generalizable Learners
  3. 2025Generalizing Vision-Language Models with Dedicated Prompt Guidance
  4. 2025Input Domain Aware MoE: Decoupling Routing Decisions from Task Optimization in Mixture of Experts
  5. 2025GMoPE:A Prompt-Expert Mixture Framework for Graph Foundation Models
  6. 2024PracticalDG: Perturbation Distillation on Vision-Language Models for Hybrid Domain Generalization
  7. 2026STAR: Rethinking MoE Routing as Structure-Aware Subspace Learning
  8. 2026L2R: Low-Rank and Lipschitz-Controlled Routing for Mixture-of-Experts