Research map / Vision-language and multimodal / Cross-modal learning
A Comprehensive Survey on Multimodal Recommender Systems: Taxonomy, Evaluation, and Future Directions
Hongyu Zhou, Xin Zhou, Zhiwei Zeng, Lingzi Zhang, Chunyan Miao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use multimodal content features instead of interactions only (VBPR vs BPR) | BPR · Baby | Recall@20: 0.0575 -> 0.0663 | Improved |
| use LightGCN instead of early multimodal MMGCN | MMGCN · Baby | Recall@20: 0.0615 -> 0.0754 | Improved |
| adopt LightGCN-style propagation in modality graphs (DualGNN over MMGCN) | MMGCN · Baby | Recall@20: 0.0615 -> 0.0716 | Improved |
| add frozen item-item graph with degree-sensitive edge pruning (FREEDOM over LATTICE) | LATTICE · Baby | Recall@20: 0.0850 -> 0.0992 | Improved |
| use FREEDOM on a small graph versus on large Electronics graph | FREEDOM · Elec | Recall@20: 0.0992 -> 0.0601 | Hurt |
| use multimodal methods instead of CF on FoodRec | LightGCN · FoodRec | Recall@20: 0.0546 -> 0.0578 | No clear effect |
| use LATTICE item graph on large graph | LATTICE · Elec | Recall@20 | Hurt |
| use ADDVAE on noisy text features of FoodRec | ADDVAE · FoodRec | Recall@20: 0.091 -> 0.0508 | Hurt |
| add self-supervised dropout contrastive learning (BM3 over LATTICE) | LATTICE · Sports | Recall@20: 0.0953 -> 0.0980 | Improved |
Closest papers by meaning
- Harnessing Multimodal Large Language Models for Multimodal Sequential Recommendation (AAAI 2025 · similarity 0.90)
- DOGE: LLMs-Enhanced Hyper-Knowledge Graph Recommender for Multimodal Recommendation (AAAI 2025 · similarity 0.89)
- MENTOR: Multi-level Self-supervised Learning for Multimodal Recommendation (AAAI 2025 · similarity 0.89)
- LGMRec: Local and Global Graph Learning for Multimodal Recommendation (AAAI 2024 · similarity 0.87)
- Seeing Beyond Noise: Joint Graph Structure Evaluation and Denoising for Multimodal Recommendation (AAAI 2025 · similarity 0.87)
- Towards Unified Multi-Modal Personalization: Large Vision-Language Models for Generative Recommendation and Beyond (ICLR 2024 · similarity 0.87)
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.
