Beyond the Aggregation Dilemma: Prior-Retaining Decoupled Learning for Multimodal Graphs

• URL: http://arxiv.org/abs/2605.24684v1
• Date: Sat, 23 May 2026 17:42:25 GMT
• Title: Beyond the Aggregation Dilemma: Prior-Retaining Decoupled Learning for Multimodal Graphs
• Abstract summary: MultimodalAttributed Graph Learning (MAGL) integrates node attributes with structural topology via graph aggregation.<n>As pretrained encoders evolve into Large Foundation Models (LFMs), the landscape of MAGL fundamentally shifts.<n>Under high-confidence LFM priors, mandatory aggregation introduces topological intrinsic noise that overwhelms discriminative signals.<n>We propose SUPRA (Shared-Unique Prior-Retaining Architecture), a decoupled dual-pathway paradigm.<n> SUPRA processes modality-specific features through topology-agnostics while capturing structural synergy via a lightweight shared GNN, with auxiliary deep supervision counteracting gradient starvation.
• Score: 69.56407708794845
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multimodal Attributed Graph Learning (MAGL) integrates intrinsic node attributes with structural topology via graph aggregation. However, as pretrained encoders evolve into Large Foundation Models (LFMs), the landscape of MAGL fundamentally shifts: under high-confidence LFM priors, mandatory aggregation introduces topological noise that overwhelms discriminative signals, triggering a counter-intuitive performance inversion where sophisticated MAGL architectures underperform simple topology-agnostic MLPs. Through systematic empirical and theoretical analysis, we identify that this inversion stems from a fundamental aggregation dilemma characterized by two concurrent pathologies: (1) Representational Pathology (SNR Degradation) - mandatory aggregation dilutes robust intrinsic features with topological noise, causing the noise penalty to outweigh its collaborative benefit; and (2) Optimization Pathology (Gradient Starvation) - topological aggregation attenuates gradient flow, while a shared task loss causes dominant modalities to prematurely suppress weaker ones. To resolve this dilemma, we propose SUPRA (Shared-Unique Prior-Retaining Architecture), a decoupled dual-pathway paradigm. SUPRA processes modality-specific features through topology-agnostic MLPs while capturing structural synergy via a lightweight shared GNN, with auxiliary deep supervision counteracting gradient starvation. Extensive evaluations demonstrate that SUPRA achieves state-of-the-art performance while requiring 3.5x lower peak GPU memory and up to 4.4x faster training time than Multimodal Graph Transformers.

Related papers

• PhyG-MoE: A Physics-Guided Mixture-of-Experts Framework for Energy-Efficient GNSS Interference Recognition [49.955269674859004]
  This paper introduces PhyG-MoE (Physics-Guided Mixture-of-Experts), a framework designed to align model capacity with signal complexity.<n>Unlike static architectures, the proposed system employs a spectrum-based gating mechanism that routes signals based on their spectral feature entanglement.<n>A high-capacity TransNeXt expert is activated on-demand to disentangle complex features in saturated scenarios, while lightweight experts handle fundamental signals to minimize latency.
  arXiv  Detail & Related papers  (2026-01-19T07:57:52Z)
• Signal-SGN++: Topology-Enhanced Time-Frequency Spiking Graph Network for Skeleton-Based Action Recognition [0.23332469289621785]
  Spiking Neural Networks (SNNs) offer energy efficiency but remain limited in capturing temporal-frequency and topological dependencies of human motion.<n>This article proposes Signal-SGN++, a topology-aware spiking graph framework that integrates adaptivity with time-frequency spiking dynamics.
  arXiv  Detail & Related papers  (2025-12-22T09:16:04Z)
• Scalable Quantum Walk-Based Heuristics for the Minimum Vertex Cover Problem [0.0]
  We propose a novel quantum algorithm for the Minimum Vertex Cover (MVC) problem based on continuous-time quantum walks (CTQWs)<n>In this framework, the coherent propagation of a quantum walker over a graph encodes its structural properties into state amplitudes.<n>We show that the CTQW-based algorithm consistently achieves superior approximation ratios and exhibits remarkable robustness with respect to network topology.
  arXiv  Detail & Related papers  (2025-12-02T17:04:57Z)
• Decoupling and Damping: Structurally-Regularized Gradient Matching for Multimodal Graph Condensation [3.2987327415317895]
  We propose Structurally-Regularized Gradient Matching (SR-GM), a novel condensation framework tailored for multimodal graphs.<n> SR-GM significantly improves accuracy and accelerates convergence compared to baseline methods.<n>This research provides a scalable methodology for multimodal graph-based learning in resource-constrained environments.
  arXiv  Detail & Related papers  (2025-11-25T11:50:34Z)
• Moving Beyond Diffusion: Hierarchy-to-Hierarchy Autoregression for fMRI-to-Image Reconstruction [65.67001243986981]
  We propose MindHier, a coarse-to-fine fMRI-to-image reconstruction framework built on scale-wise autoregressive modeling.<n>MindHier achieves superior semantic fidelity, 4.67x faster inference, and more deterministic results than the diffusion-based baselines.
  arXiv  Detail & Related papers  (2025-10-25T15:40:07Z)
• Learning Efficient and Generalizable Graph Retriever for Knowledge-Graph Question Answering [75.12322966980003]
  Large Language Models (LLMs) have shown strong inductive reasoning ability across various domains.<n>Most existing RAG pipelines rely on unstructured text, limiting interpretability and structured reasoning.<n>Recent studies have explored integrating knowledge graphs with LLMs for knowledge graph question answering.<n>We propose RAPL, a novel framework for efficient and effective graph retrieval in KGQA.
  arXiv  Detail & Related papers  (2025-06-11T12:03:52Z)
• Towards Anomaly-Aware Pre-Training and Fine-Tuning for Graph Anomaly Detection [59.042018542376596]
  Graph anomaly detection (GAD) has garnered increasing attention in recent years, yet remains challenging due to two key factors.<n>Anomaly-Aware Pre-Training and Fine-Tuning (APF) is a framework to mitigate the challenges in GAD.<n> Comprehensive experiments on 10 benchmark datasets validate the superior performance of APF in comparison to state-of-the-art baselines.
  arXiv  Detail & Related papers  (2025-04-19T09:57:35Z)
• Perturbation Ontology based Graph Attention Networks [26.95077612390953]
  Ontology-based Graph Attention Networks (POGAT) is a novel methodology that combines ontology subgraphs with an advanced self-supervised learning paradigm to achieve a deep contextual understanding.<n>POGAT significantly outperforms state-of-the-art baselines, achieving a groundbreaking improvement of up to 10.78% in F1-score for the critical task of link prediction and 12.01% in Micro-F1 for the critical task of node classification.
  arXiv  Detail & Related papers  (2024-11-27T17:12:14Z)
• DA-Flow: Dual Attention Normalizing Flow for Skeleton-based Video Anomaly Detection [52.74152717667157]
  We propose a lightweight module called Dual Attention Module (DAM) for capturing cross-dimension interaction relationships in-temporal skeletal data. It employs the frame attention mechanism to identify the most significant frames and the skeleton attention mechanism to capture broader relationships across fixed partitions with minimal parameters and flops.
  arXiv  Detail & Related papers  (2024-06-05T06:18:03Z)
• GA-HQS: MRI reconstruction via a generically accelerated unfolding approach [14.988694941405575]
  We propose a Generically Accelerated Half-Quadratic Splitting (GA-HQS) algorithm that incorporates second-order gradient information and pyramid attention modules for the delicate fusion of inputs at the pixel level. Our method surpasses previous ones on single-coil MRI acceleration tasks.
  arXiv  Detail & Related papers  (2023-04-06T06:21:18Z)
• Counterfactual Intervention Feature Transfer for Visible-Infrared Person Re-identification [69.45543438974963]
  We find graph-based methods in the visible-infrared person re-identification task (VI-ReID) suffer from bad generalization because of two issues. The well-trained input features weaken the learning of graph topology, making it not generalized enough during the inference process. We propose a Counterfactual Intervention Feature Transfer (CIFT) method to tackle these problems.
  arXiv  Detail & Related papers  (2022-08-01T16:15:31Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.