Well Begun is Half Done: Training-Free and Model-Agnostic Semantically Guaranteed User Representation Initialization for Multimodal Recommendation
sigir26-099 · Full research paper · Jinfeng Xu, Zheyu Chen, Shuo Yang, Jinze Li, Hewei Wang, Jianheng Tang, Wei Wang, Xiping Hu, Edith C.‐H. Ngai
Multimodal Denoising and Robustness · 100%
Abstract
Recent advancements in multimodal recommendations, which leverage diverse modality information to mitigate data sparsity and improve recommendation accuracy, have gained significant attention. However, existing multimodal recommendations overlook the critical role of user representation initialization. Unlike items, which are naturally associated with rich modality information, users lack such inherent information. Consequently, item representations initialized based on meaningful modality information and user representations initialized randomly exhibit a significant semantic gap.
Assigned reviewers
| # | Reviewer | Match | Why (closest evidence publications in the taxonomy) | |
|---|---|---|---|---|
| 1 | Xin Zhang 0091 | 41 | 100 Multi-Modal Correction Network for Recommendation (2025) Multimodal Denoising and Robustness 33 Deep Dynamic Interest Learning With Session Local and Global Consistency for Click-Through Rate Predictions (2022) CTR and Conversion Rate Prediction 33 Deep Session Heterogeneity-Aware Network for Click Through Rate Prediction (2024) CTR and Conversion Rate Prediction |
|
| 2 | Fengling Li 0001 | 34 | 49 Flip is Better than Noise: Unbiased Interest Generation for Multimedia Recommendation (2025) Diffusion and Flow Recommenders 48 SOIL: Contrastive Second-Order Interest Learning for Multimodal Recommendation (2024) Graph Contrastive and Self-Supervised Learning 39 Task-Adversarial Adaptation for Multi-modal Recommendation (2023) Domain Adaptation and Disentangled Representation for CDR |
Area chair
| Area chair | Match | |
|---|---|---|
| Weiming Liu 0005 | 32 |
Area-chair candidates (5)
| Xiaoye Qu | 32 | 15 papers | |
| Weiming Liu 0005 assigned | 32 | 8 papers | |
| Xiaoyi Zeng | 30 | 11 papers | |
| Leyu Lin | 30 | 15 papers | |
| Kun Gai | 30 | 5 papers |
Candidates from the panel ranked by match score
| # | Reviewer | Match | Load | Why | |
|---|---|---|---|---|---|
| 1 | Xin Zhang 0091 assigned | 41 | 3 | 100 Multi-Modal Correction Network for Recommendation (2025) Multimodal Denoising and Robustness 33 Deep Dynamic Interest Learning With Session Local and Global Consistency for Click-Through Rate Predictions (2022) CTR and Conversion Rate Prediction |
|
| 2 | Yinwei Wei COI: co-author | 41 | 3 | 65 Breaking Isolation: Multimodal Graph Fusion for Multimedia Recommendation by Edge-wise Modulation (2022) Graph Contrastive and Self-Supervised Learning 65 Leveraging Multimodal Features and Item-level User Feedback for Bundle Construction (2024) Graph Contrastive and Self-Supervised Learning |
|
| 3 | Haokun Wen | 39 | 3 | 41 Multimodal Compatibility Modeling via Exploring the Consistent and Complementary Correlations (2021) Universal Multimodal Embedding and MLLM Retrieval 41 Attribute-wise Explainable Fashion Compatibility Modeling (2021) Universal Multimodal Embedding and MLLM Retrieval |
|
| 4 | Hao Deng 0011 | 38 | 3 | 65 ESANS: Effective and Semantic-Aware Negative Sampling for Large-Scale Retrieval Systems (2025) Multimodal Denoising and Robustness 39 MMQ: Multimodal Mixture-of-Quantization Tokenization for Semantic ID Generation and User Behavioral Adaptation (2026) Item Tokenization and Semantic IDs |
|
| 5 | Minghui Fang 0002 | 38 | 3 | 41 OmniSep: Unified Omni-Modality Sound Separation with Query-Mixup (2025) Multimodal Retrieval and Embedding 41 ControlSpeech: Towards Simultaneous and Independent Zero-shot Speaker Cloning and Zero-shot Language Style Control (2025) Multimodal Retrieval and Embedding |
|
| 6 | Yongxin Ni | 36 | 3 | 56 Teach Me How to Denoise: A Universal Framework for Denoising Multi-modal Recommender Systems via Guided Calibration (2025) Multimodal Generative Recommendation 44 Causality-Inspired Fair Representation Learning for Multimodal Recommendation (2025) Bias Mitigation Methods |
|
| 7 | Jian Lang | 34 | 3 | 41 REDEEMing Modality Information Loss: Retrieval-Guided Conditional Generation for Severely Modality Missing Learning (2025) Universal Multimodal Embedding and MLLM Retrieval 41 Seeing the Unseen in Micro-Video Popularity Prediction: Self-Correlation Retrieval for Missing Modality Generation (2025) Universal Multimodal Embedding and MLLM Retrieval |
|
| 8 | Haiyang Xu 0001 COI: co-author | 34 | 3 | 41 Efficient and Effective In-context Demonstration Selection with Coreset (2026) Multimodal Retrieval and Embedding 41 Vision Language Pre-training by Contrastive Learning with Cross-Modal Similarity Regulation (2023) Multimodal Retrieval and Embedding |
|
| 9 | Jiuqiang Li 0002 | 34 | 3 | 54 Multi-view Semantic Contrastive Alignment for Multimodal Recommendation (2026) Universal Multimodal Embedding and MLLM Retrieval 26 Disentangled Contrastive Bundle Recommendation with Conditional Diffusion (2025) Diffusion and Flow Recommenders |
|
| 10 | Fengling Li 0001 assigned | 34 | 3 | 49 Flip is Better than Noise: Unbiased Interest Generation for Multimedia Recommendation (2025) Diffusion and Flow Recommenders 48 SOIL: Contrastive Second-Order Interest Learning for Multimodal Recommendation (2024) Graph Contrastive and Self-Supervised Learning |
|
| 11 | Jinpeng Wang 0002 | 34 | 3 | 41 Imagine with Layout and Sketch: Enhancing Vision-Language Retrieval with Dual-Stream Multi-Modal Query Refinement (2026) Universal Multimodal Embedding and MLLM Retrieval 41 Suit the Remedy to the Retriever: Interpretable Query Optimization with Retriever Preference Alignment for Vision-Language Retrieval (2026) Universal Multimodal Embedding and MLLM Retrieval |
|
| 12 | Ke Fei COI: co-author | 34 | 3 | 37 Sharpness-Consistent Cross-Domain Recommendation for Cold-Start Items (2026) Item Tokenization and Semantic IDs 35 Entire-Space Variational Information Exploitation for Post-Click Conversion Rate Prediction (2025) CTR and Conversion Rate Prediction |
|
| 13 | Zihao Yue | 34 | 3 | 39 Learning Descriptive Image Captioning via Semipermeable Maximum Likelihood Estimation (2023) Multimodal Evaluation and Benchmarks 39 Less is More: Mitigating Multimodal Hallucination from an EOS Decision Perspective (2024) Multimodal Evaluation and Benchmarks |
|
| 14 | Junchen Fu | 33 | 3 | 56 Teach Me How to Denoise: A Universal Framework for Denoising Multi-modal Recommender Systems via Guided Calibration (2025) Multimodal Generative Recommendation 43 Are Multimodal Embeddings Truly Beneficial for Recommendation? A Deep Dive into Whole vs. Individual Modalities (2026) Multimodal Generative Recommendation |
|
| 15 | Miaomiao Cai 0001 | 33 | 3 | 70 I3-MRec: Invariant Learning with Information Bottleneck for Incomplete Modality Recommendation (2025) Multimodal Denoising and Robustness 52 Multimodality Invariant Learning for Multimedia-Based New Item Recommendation (2024) Content and Attribute-Based Cold-Start Recommendation |
|
| 16 | Zheng Zhang 0006 | 33 | 3 | 33 Work Together: Correlation-Identity Reconstruction Hashing for Unsupervised Cross-Modal Retrieval (2023) Cross-Modal Hashing and Quantization 33 Cross-Domain Transfer Hashing for Efficient Cross-Modal Retrieval (2024) Cross-Modal Hashing and Quantization |
|
| 17 | Jiangxia Cao | 33 | 3 | 57 QARM: Quantitative Alignment Multi-Modal Recommendation at Kuaishou (2025) Multimodal Generative Recommendation 40 A Multi-modal Modeling Framework for Cold-start Short-video Recommendation (2024) Multimodal Generative Recommendation |
|
| 18 | Qijiong Liu | 32 | 3 | 43 Multimodal Pretraining, Adaptation, and Generation for Recommendation: A Survey (2024) Multimodal Generative Recommendation 39 EasyGen: Easing Multimodal Generation with BiDiffuser and LLMs (2024) Multimodal Evaluation and Benchmarks |
|
| 19 | Jinxin Hu | 32 | 3 | 64 ESANS: Effective and Semantic-Aware Negative Sampling for Large-Scale Retrieval Systems (2025) Multimodal Denoising and Robustness 39 MMQ: Multimodal Mixture-of-Quantization Tokenization for Semantic ID Generation and User Behavioral Adaptation (2026) Item Tokenization and Semantic IDs |
|
| 20 | Yong He 0009 | 32 | 3 | 36 Uncertainty-based Heterogeneous Privileged Knowledge Distillation for Recommendation System (2023) Domain Adaptation and Disentangled Representation for CDR 36 The Devil is in the Sources! Knowledge Enhanced Cross-Domain Recommendation in an Information Bottleneck Perspective (2024) Domain Adaptation and Disentangled Representation for CDR |
|
| 23 | Lichan Hong | 31 | 3 | 42 Enhancing Online Ranking Systems via Multi-Surface Co-Training for Content Understanding (2025) Content-Based Transferable Recommendation 35 DCN V2: Improved Deep & Cross Network and Practical Lessons for Web-scale Learning to Rank Systems (2021) CTR and Conversion Rate Prediction |
|
| 24 | Meiyu Liang | 31 | 3 | 41 Dynamic Self-adaptive Multiscale Distillation from Pre-trained Multimodal Large Model for Efficient Cross-modal Retrieval (2025) Universal Multimodal Embedding and MLLM Retrieval 41 Knowledge Graph Enhanced Multimodal Transformer for Image-Text Retrieval (2024) Universal Multimodal Embedding and MLLM Retrieval |
|
| 25 | Linchao Zhu | 31 | 3 | 41 Knowledge-Enhanced Dual-Stream Zero-Shot Composed Image Retrieval (2024) Universal Multimodal Embedding and MLLM Retrieval 41 Attention as Selector: Unlocking VLM Attention for Long Document Page Retrieval (2026) Universal Multimodal Embedding and MLLM Retrieval |
|
| 26 | Junwei Pan | 31 | 3 | 35 Long-Sequence Recommendation Models Need Decoupled Embeddings (2025) CTR and Conversion Rate Prediction 35 From Feature Interaction to Feature Generation: A Generative Paradigm of CTR Prediction Models (2025) CTR and Conversion Rate Prediction |
|
| 27 | Yanzhao Zhang | 31 | 3 | 41 Bridging Modalities: Improving Universal Multimodal Retrieval by Multimodal Large Language Models (2025) Universal Multimodal Embedding and MLLM Retrieval 41 Breaking the Modality Barrier: Universal Embedding Learning with Multimodal LLMs (2025) Universal Multimodal Embedding and MLLM Retrieval |
|
| 28 | Jieming Zhu | 30 | 3 | 39 Directional Adversarial Training for Recommender Systems (2020) Robustness to Noise and Distribution Shift 35 UNGER: Generative Recommendation with A Unified Code via Semantic and Collaborative Integration (2026) Item Tokenization and Semantic IDs |
|
| 30 | Qiang Luo 0004 | 30 | 3 | 39 MISS: Multi-Modal Tree Indexing and Searching with Lifelong Sequential Behavior for Retrieval Recommendation (2025) Universal Multimodal Embedding and MLLM Retrieval 36 FARM: Frequency-Aware Model for Cross-Domain Live-Streaming Recommendation (2026) Domain Adaptation and Disentangled Representation for CDR |
|
| 32 | Ke Wang 0038 | 30 | 3 | 35 Contrastive Self-supervised Learning in Recommender Systems: A Survey (2024) Graph Contrastive and Self-Supervised Learning 34 Review-Enhanced Hierarchical Contrastive Learning for Recommendation (2024) Graph Contrastive and Self-Supervised Learning |
|
| 33 | Riwei Lai | 30 | 3 | 41 WISE: Word-Level Interaction-Based Multimodal Fusion for Speech Emotion Recognition (2020) Multimodal Retrieval and Embedding 39 DAR: Dimension-Adaptive Recommendation with Multi-Granular Noise Control (2025) Robustness to Noise and Distribution Shift |
|
| 34 | Yi Zhang 0103 | 30 | 3 | 48 Intent Representation Learning with Large Language Model for Recommendation (2025) Collaborative-Semantic Representation Alignment 35 SimCEN: Simple Contrast-enhanced Network for CTR Prediction (2024) CTR and Conversion Rate Prediction |