TMTE: Effective Multimodal Graph Learning with Task-aware Modality and Topology Co-evolution

TL;DR

TMTE introduces a novel framework for multimodal graph learning that jointly optimizes graph topology and modality representations, addressing real-world MAG limitations and improving task performance.

Contribution

It proposes a task-aware co-evolution approach for topology and modality, with a bidirectional coupling mechanism and a closed-loop optimization process.

Findings

Achieves state-of-the-art results on 9 MAG datasets and 1 non-graph dataset.

Effectively addresses noise and missing connections in real-world MAGs.

Demonstrates consistent improvements across diverse graph-centric and modality-centric tasks.

Abstract

Multimodal-attributed graphs (MAGs) are a fundamental data structure for multimodal graph learning (MGL), enabling both graph-centric and modality-centric tasks. However, our empirical analysis reveals inherent topology quality limitations in real-world MAGs, including noisy interactions, missing connections, and task-agnostic relational structures. A single graph derived from generic relationships is therefore unlikely to be universally optimal for diverse downstream tasks. To address this challenge, we propose Task-aware Modality and Topology co-Evolution (TMTE), a novel MGL framework that jointly and iteratively optimizes graph topology and multimodal representations toward the target task. TMTE is motivated by the bidirectional coupling between modality and topology: multimodal attributes induce relational structures, while graph topology shapes modality representations. Concretely, TMTE casts topology evolution as multi-perspective metric learning over modality embeddings with an anchor-based approximation, and formulates modality evolution as smoothness-regularized fusion with cross-modal alignment, yielding a closed-loop task-aware co-evolution process. Extensive experiments on 9 MAG datasets and 1 non-graph multimodal dataset across 6 graph-centric and modality-centric tasks show that TMTE consistently achieves state-of-the-art performance. Our code is available at https://anonymous.4open.science/r/TMTE-1873.