This work proposes TEAM, a modality-centric framework that constructs and encodes graph context with node-modality tokens rather than treating each node as a single interaction unit, and introduces the Modality-Source Context Transformer (MSCT), which encodes the selected node-modality tokens and coordinates interactions among modality-source groups.
Abstract
Multimodal-Attributed Graphs (MAGs) organize heterogeneous data in graphs whose nodes are characterized by modalities such as text and images rather than traditional unimodal attributes. However, many representative MAG methods organize propagation around node-level states and topology-defined neighborhoods, which can obscure fine-grained modality contributions: a bounded neighborhood may contain semantically weak evidence, while compatible context need not be confined to it. To address both limitations, we propose TEAM, a modality-centric framework that constructs and encodes graph context with node-modality tokens rather than treating each node as a single interaction unit. Concretely, TEAM unfolds the local topology into a Modality Encoding Tree and performs tree-anchored semantic evidence acquisition, organizing local structural context while expanding the accessible evidence space outside the constructed local Tree. We further introduce the Modality-Source Context Transformer (MSCT), which encodes the selected node-modality tokens and coordinates interactions among modality-source groups. By retaining textual and visual attributes as separate interaction units until readout, it supports fine-grained cross-node cross-modal exchange without first collapsing them into whole-node features. Across the evaluated baselines, TEAM attains the highest reported mean on all graph-learning metrics, all three G2Text metrics, and both G2Image metrics. On Toys, it surpasses the strongest baseline by 3.37 and 2.93 percentage points in Accuracy and Macro-F1, respectively.
GraphBind is a topology-driven approach that uses graph topology to bind rich modality information into a unified shared space that integrates structure and semantics, and adapts this space to discriminative and generative tasks through lightweight interfaces.
Xun-Kai Li, Chen-Xi Wan, Yin-Lin Zhu et al.· 0 citations
This work introduces ICE (Interaction-aware Clifford Encoder), a multimodal graph foundation model built on a node-indexed Clifford latent field that ranks first in all 30 reported supervised and few-shot comparisons.
Community Search (CS) aims to retrieve query-dependent cohesive subgraphs from networks, playing a crucial role in modern applications such as computing power networks and biological systems. Given the high cost of labeling, unsupervised CS has emerged as a scalable paradigm. However, existing approaches often suffer f...
Xi Yue, Chuan-Hou Sun, Ying Yin et al.· Proceedings of the 35th ACM...· 0 citations
Multimodal graphs integrate textual and visual attributes with relational structures, providing a rich representation of complex real-world data. To enable the broad applicability and reuse of graph models, it is crucial for models trained on source graphs to generalize to unseen graph domains. However, this is challen...
Hai-Yang Jiang, Tong Chen, Xin Gao et al.· 0 citations
Cross-modal retrieval has made remarkable strides with embedding-based methods that map different modalities into a unified space. However, these approaches fail on compositional queries that require understanding multiple entities and their relationships. The root cause is the flat feature paradigm, which compresses s...
Xin-Biao Gan, Hong-Xu Jin· Proceedings of the 34th ACM...· 0 citations
Multimodal recommender systems leverage heterogeneous content such as images and text to enhance user preference modeling. However, existing methods suffer from two fundamental limitations: (1) they inadequately model the compositional nature of multimodal semantics, often overemphasizing modality-shared information wh...
De-Sheng Cai, Ying He, Sheng-Sheng Qian et al.· Proceedings of the 34th ACM...· 0 citations
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduNov 17, 2026
By studying the computational foundations of language learning, Associate Professor Jacob Andreas builds intelligent systems that can communicate effectively with humans.
Microsoft Research Blog· microsoft.comNov 16, 2026
AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.
MIT News · Artificial Intelligence· news.mit.eduNov 13, 2026