AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

MM-DyGraph: A Dataset, Benchmark, and Model for Multimodal Dynamic Graphs

Zongyuan Wu, Pengjie Wang, Lingshan Chen, Tianhang Wan, Yuan Meng, Xin Wang, Ling Feng, Wenwu Zhu

NeurIPS 2026 · Vision-language and multimodal › Cross-modal learning · approach: Multimodal sentiment analysis · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Introduces MM-DyGraph, a dataset, benchmark, and model for multimodal dynamic graphs where nodes and edges carry text and image signals.

Problem

Benchmarks and methods explicitly studying multimodal learning on dynamic graphs (nodes and edges with text and images) are scarce, limiting use of multimodal information for downstream tasks.

Approach

Provides a multimodal dynamic graph dataset and benchmark together with a proposed model for learning over text- and image-attributed temporal graphs; specific method details are not stated in the available abstract.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add text embeddings instead of random featuresTST-MF · GDELT LP transductiveAP (%): 92.83 -> 94.03Improved
add image embeddings on top of text (naive concatenation)DyGFormer · Amazon-Office LP inductiveAP (%): 73.50 -> 68.64Hurt
add image embeddings on top of textTST-MF · Amazon-Office LP inductiveAP (%): 77.47 -> 78.16Improved
add image embeddings on top of textTST-MF · Yelp node retrieval transductiveHits@10 (%): 75.19 -> 79.11Improved
add image embeddings on top of textCAWN · Amazon-Industrial LP transductiveAP (%): 91.94 -> 92.21Improved
replace naive concatenation with target-aware fusionDyGFormer · Amazon-Industrial LP inductive (Text+Image)AP (%): 76.33 -> 81.95Improved
replace naive concatenation with target-aware fusionDyGFormer · GMCT LP transductive (Text+Image)AP (%): 95.54 -> 89.08Hurt
use target-aware fusion instead of gated fusionDyGFormer backbone · GDELT node retrieval transductiveHits@10 (%): 62.18 -> 81.62Improved
use target-aware fusion instead of modality attentionDyGFormer backbone · Yelp node retrieval transductiveHits@10 (%): 57.09 -> 79.11Improved
use target-aware fusion instead of naive concatenationDyGFormer · Amazon-Industrial edge classificationPrecision (%): 59.38 -> 50.32Hurt
add visual modality with misaligned source (edge/node mismatch)dynamic graph methods · MM-DyGraphvariousMixed
add multimodal features on saturated structural datasetsstrong backbones · GDELT / Amazon-Office link predictionAP/AUCNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy