DIGEST: Dynamic Graph Refinement with Dual Contrastive Semantic Transfer for Multimodal Recommendation
sigir26-118 · Full research paper · Xiangyu Sai, Meysam Madadi, Sérgio Escalera, Yong Xu
Graph Contrastive and Self-Supervise · 50% Graph Contrastive and Self-Supervise · 25% Multimodal Denoising and Robustness · 25%
Abstract
Multimodal recommendation benefits from leveraging rich content signals such as images and texts to alleviate interaction sparsity, yet existing graph-based approaches are still hindered by (i) noisy user—item edges that are treated as static during training and (ii) inconsistent representation spaces across interaction-driven and modality-induced graph views. To address these issues, we propose DIGEST, a multi-graph framework that propagates trainable ID embeddings on a denoised user—item graph and a fused modality-induced item—item graph, and interleaves message passing with dynamic graph refinement that iteratively reweights existing edges to suppress noisy connections. To enable reliable semantic transfer across views, DIGEST further introduces a dual contrastive alignment that (i) aligns the collaborative and semantic item views and (ii) constrains the semantic graph representations to projected multimodal features, together with a lightweight dimension decorrelation regularizer and adaptive gated fusion to reduce redundancy and stabilize multi-view learning. Extensive experiments on three Amazon benchmark datasets demonstrate that DIGEST consistently outperforms state-of-the-art multimodal recommenders, achieving up to 8.43% relative improvement on NDCG@20 and 7.66% on Recall@20 over the strongest baselines.
Assigned reviewers
| # | Reviewer | Match | Why (closest evidence publications in the taxonomy) | |
|---|---|---|---|---|
| 1 | Yinwei Wei | 74 | 100 Modality-Independent Graph Neural Networks with Global Transformers for Multimodal Recommendation (2025) Graph Contrastive and Self-Supervised Learning 100 Multi-modal Bipartite Graph Structure Learning with Information Bottleneck for Micro-video Recommendation (2026) Graph Contrastive and Self-Supervised Learning 93 Graph-Refined Convolutional Network for Multimedia Recommendation with Implicit Feedback (2020) Graph Contrastive and Self-Supervised Learning |
|
| 2 | Jiuqiang Li 0002 | 60 | 62 Multi-view Semantic Contrastive Alignment for Multimodal Recommendation (2026) Universal Multimodal Embedding and MLLM Retrieval 59 Graph Diffusive Self-Supervised Learning for Social Recommendation (2024) Social and Group Recommendation 47 Disentangled Contrastive Bundle Recommendation with Conditional Diffusion (2025) Diffusion and Flow Recommenders |
|
| 3 | Yi Zhang 0103 | 56 | 84 MixRec: Individual and Collective Mixing Empowers Data Augmentation for Recommender Systems (2025) Graph Contrastive and Self-Supervised Learning 84 Unveiling Contrastive Learning's Capability of Neighborhood Aggregation for Collaborative Filtering (2025) Graph Contrastive and Self-Supervised Learning 69 Intent-Guided Heterogeneous Graph Contrastive Learning for Recommendation (2025) Graph Contrastive and Self-Supervised Learning |
Area chair
| Area chair | Match | |
|---|---|---|
| Leyu Lin | 46 |
Area-chair candidates (4)
| Leyu Lin assigned | 46 | 15 papers | |
| Weiming Liu 0005 | 44 | 8 papers | |
| Qiaoyu Tan | 43 | 7 papers | |
| Kun Gai | 39 | 5 papers |
Candidates from the panel ranked by match score
| # | Reviewer | Match | Load | Why | |
|---|---|---|---|---|---|
| 1 | Yinwei Wei assigned | 74 | 3 | 100 Modality-Independent Graph Neural Networks with Global Transformers for Multimodal Recommendation (2025) Graph Contrastive and Self-Supervised Learning 100 Multi-modal Bipartite Graph Structure Learning with Information Bottleneck for Micro-video Recommendation (2026) Graph Contrastive and Self-Supervised Learning |
|
| 2 | Lianghao Xia COI: co-author | 62 | 3 | 84 LightGCL: Simple Yet Effective Graph Contrastive Learning for Recommendation (2023) Graph Contrastive and Self-Supervised Learning 84 Automated Self-Supervised Learning for Recommendation (2023) Graph Contrastive and Self-Supervised Learning |
|
| 3 | Ke Wang 0038 | 60 | 3 | 85 Review-Enhanced Hierarchical Contrastive Learning for Recommendation (2024) Graph Contrastive and Self-Supervised Learning 85 Multi-aspect Graph Contrastive Learning for Review-enhanced Recommendation (2024) Graph Contrastive and Self-Supervised Learning |
|
| 4 | Jiuqiang Li 0002 assigned | 60 | 3 | 62 Multi-view Semantic Contrastive Alignment for Multimodal Recommendation (2026) Universal Multimodal Embedding and MLLM Retrieval 59 Graph Diffusive Self-Supervised Learning for Social Recommendation (2024) Social and Group Recommendation |
|
| 5 | Mengyuan Jing | 60 | 3 | 70 Review-Enhanced Hierarchical Contrastive Learning for Recommendation (2024) Graph Contrastive and Self-Supervised Learning 70 Graph Contrastive Learning with Adaptive Augmentation for Recommendation (2022) Graph Contrastive and Self-Supervised Learning |
|
| 6 | Yi Zhang 0103 assigned | 56 | 3 | 84 MixRec: Individual and Collective Mixing Empowers Data Augmentation for Recommender Systems (2025) Graph Contrastive and Self-Supervised Learning 84 Unveiling Contrastive Learning's Capability of Neighborhood Aggregation for Collaborative Filtering (2025) Graph Contrastive and Self-Supervised Learning |
|
| 7 | Xubin Ren COI: co-author | 55 | 3 | 70 LightGCL: Simple Yet Effective Graph Contrastive Learning for Recommendation (2023) Graph Contrastive and Self-Supervised Learning 70 SSLRec: A Self-Supervised Learning Framework for Recommendation (2024) Graph Contrastive and Self-Supervised Learning |
|
| 8 | Dongxiao He | 52 | 3 | 70 A New Mechanism for Eliminating Implicit Conflict in Graph Contrastive Learning (2024) Graph Contrastive and Self-Supervised Learning 70 Self-Supervised Graph Neural Networks via Diverse and Interactive Message Passing (2022) Graph Contrastive and Self-Supervised Learning |
|
| 9 | Wei Wei 0027 COI: co-author | 52 | 3 | 69 Multi-Modal Self-Supervised Learning for Recommendation (2023) Graph Contrastive and Self-Supervised Learning 69 Heterogeneous Graph Contrastive Learning for Recommendation (2023) Graph Contrastive and Self-Supervised Learning |
|
| 10 | Chunyang Wang 0001 | 51 | 3 | 100 Guiding Graph Learning with Denoised Modality for Multi-modal Recommendation (2024) Graph Contrastive and Self-Supervised Learning 85 Review-Enhanced Hierarchical Contrastive Learning for Recommendation (2024) Graph Contrastive and Self-Supervised Learning |
|
| 11 | Xin Zhang 0091 | 48 | 3 | 48 Multi-Modal Correction Network for Recommendation (2025) Multimodal Denoising and Robustness 43 Deep Dynamic Interest Learning With Session Local and Global Consistency for Click-Through Rate Predictions (2022) CTR and Conversion Rate Prediction |
|
| 12 | Yongjing Hao | 46 | 3 | 83 LMACL: Improving Graph Collaborative Filtering with Learnable Model Augmentation Contrastive Learning (2024) Graph Contrastive and Self-Supervised Learning 83 Meta-optimized Structural and Semantic Contrastive Learning for Graph Collaborative Filtering (2024) Graph Contrastive and Self-Supervised Learning |
|
| 14 | Miaomiao Cai 0001 | 45 | 3 | 54 I3-MRec: Invariant Learning with Information Bottleneck for Incomplete Modality Recommendation (2025) Multimodal Denoising and Robustness 54 Multimodality Invariant Learning for Multimedia-Based New Item Recommendation (2024) Content and Attribute-Based Cold-Start Recommendation |
|
| 15 | Xiaolong Liu 0012 | 45 | 3 | 60 Graph-based Alignment and Uniformity for Recommendation (2023) Graph-based User and Item Representation Learning 59 Group Identification via Transitional Hypergraph Convolution with Cross-view Self-supervised Learning (2023) Social and Group Recommendation |
|
| 16 | Lantao Hu | 45 | 3 | 70 Towards Robust Recommendation via Decision Boundary-aware Graph Contrastive Learning (2024) Graph Contrastive and Self-Supervised Learning 58 Soft Contrastive Sequential Recommendation (2024) Contrastive and Self-Supervised Sequential Recommendation |
|
| 17 | Hao Deng 0011 | 45 | 3 | 50 ESANS: Effective and Semantic-Aware Negative Sampling for Large-Scale Retrieval Systems (2025) Multimodal Denoising and Robustness 44 MMQ: Multimodal Mixture-of-Quantization Tokenization for Semantic ID Generation and User Behavioral Adaptation (2026) Item Tokenization and Semantic IDs |
|
| 19 | Bowen Zheng 0005 | 44 | 3 | 84 Enhancing Graph Contrastive Learning with Reliable and Informative Augmentation for Recommendation (2025) Graph Contrastive and Self-Supervised Learning 43 Generative Recommender with End-to-End Learnable Item Tokenization (2025) Item Tokenization and Semantic IDs |
|
| 20 | Chen Wang 0052 | 43 | 3 | 60 Graph-based Alignment and Uniformity for Recommendation (2023) Graph-based User and Item Representation Learning 48 Group Identification via Transitional Hypergraph Convolution with Cross-view Self-supervised Learning (2023) Social and Group Recommendation |
|
| 21 | Jiancan Wu | 43 | 3 | 84 Self-supervised Graph Learning for Recommendation (2021) Graph Contrastive and Self-Supervised Learning 82 Graph Contrastive Learning With Negative Propagation for Recommendation (2024) Graph Contrastive and Self-Supervised Learning |
|
| 23 | Yong He 0009 | 43 | 3 | 60 Global-Aware Model-Free Self-distillation for Recommendation System (2023) Graph Contrastive and Self-Supervised Learning 44 The Devil is in the Sources! Knowledge Enhanced Cross-Domain Recommendation in an Information Bottleneck Perspective (2024) Domain Adaptation and Disentangled Representation for CDR |
|
| 24 | Min Hou 0004 | 43 | 3 | 54 Unified Representation Learning for Discrete Attribute Enhanced Completely Cold-Start Recommendation (2025) Graph Contrastive and Self-Supervised Learning 54 Multimodality Invariant Learning for Multimedia-Based New Item Recommendation (2024) Content and Attribute-Based Cold-Start Recommendation |
|
| 25 | Yixin Liu 0001 | 42 | 3 | 70 Learning Strong Graph Neural Networks with Weak Information (2023) Graph Contrastive and Self-Supervised Learning 70 Beyond Smoothing: Unsupervised Graph Representation Learning with Edge Heterophily Discriminating (2023) Graph Contrastive and Self-Supervised Learning |
|
| 26 | Jiangxia Cao | 42 | 3 | 58 QARM: Quantitative Alignment Multi-Modal Recommendation at Kuaishou (2025) Multimodal Generative Recommendation 44 DisenCDR: Learning Disentangled Representations for Cross-Domain Recommendation (2022) Domain Adaptation and Disentangled Representation for CDR |
|
| 27 | Mingdai Yang | 42 | 3 | 60 Graph-based Alignment and Uniformity for Recommendation (2023) Graph-based User and Item Representation Learning 59 Group Identification via Transitional Hypergraph Convolution with Cross-view Self-supervised Learning (2023) Social and Group Recommendation |
|
| 28 | Lingling Yi | 42 | 3 | 72 Dual Interests-Aligned Graph Auto-Encoders for Cross-domain Recommendation in WeChat (2023) Graph Contrastive and Self-Supervised Learning 67 Item-Provider Co-learning for Sequential Recommendation (2022) Graph Contrastive and Self-Supervised Learning |
|
| 29 | Lichan Hong | 41 | 3 | 60 Self-supervised Learning for Large-scale Item Recommendations (2021) Graph Contrastive and Self-Supervised Learning 44 Better Generalization with Semantic IDs: A Case Study in Ranking for Recommendations (2024) Item Tokenization and Semantic IDs |
|
| 30 | Yongxin Ni | 41 | 3 | 55 Teach Me How to Denoise: A Universal Framework for Denoising Multi-modal Recommender Systems via Guided Calibration (2025) Multimodal Generative Recommendation 46 Causality-Inspired Fair Representation Learning for Multimodal Recommendation (2025) Bias Mitigation Methods |
|
| 31 | Riwei Lai | 41 | 3 | 53 Disentangled Negative Sampling for Collaborative Filtering (2023) Graph Contrastive and Self-Supervised Learning 40 DAR: Dimension-Adaptive Recommendation with Multi-Granular Noise Control (2025) Robustness to Noise and Distribution Shift |
|
| 32 | Jinxin Hu | 41 | 3 | 59 Distribution-Guided Auto-Encoder for User Multimodal Interest Cross Fusion (2025) Content-Based Transferable Recommendation 50 ESANS: Effective and Semantic-Aware Negative Sampling for Large-Scale Retrieval Systems (2025) Multimodal Denoising and Robustness |
|
| 33 | Ed H. Chi | 41 | 3 | 60 Self-supervised Learning for Large-scale Item Recommendations (2021) Graph Contrastive and Self-Supervised Learning 44 Better Generalization with Semantic IDs: A Case Study in Ranking for Recommendations (2024) Item Tokenization and Semantic IDs |