Research map / Vision-language and multimodal / Cross-modal learning
SGP4SR: Separated-Modality Guided User Preference Learning for Multimodal Sequential Recommendation
Changhong Li, Zhiqiang Guo, Guohui Li, Zhong Yang, Chuhang Hong
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 |
|---|---|---|---|
| add CGC module (co-occurrence-guided modal graph) (ablation) | SGP4SR · Clothing | Recall@10: 0.0267 -> 0.0541 | Improved |
| add CIP module (interest centers) (ablation) | SGP4SR · Clothing | Recall@10: 0.0369 -> 0.0541 | Improved |
| add CIP module (interest centers) (ablation) | SGP4SR · Baby | NDCG@10: 0.0281 -> 0.0283 | No clear effect |
| add CGC and CIP together (ablation) | SGP4SR · Office | NDCG@10: 0.0787 -> 0.0876 | Improved |
| add co-occurrence guidance to graph construction (ablation) | SGP4SR · Clothing | Recall@10: 0.0529 -> 0.0541 | Improved |
| add co-occurrence guidance to graph construction (ablation) | SGP4SR · Office | NDCG@10: 0.0855 -> 0.0876 | No clear effect |
| use separated-modality framework instead of fused modalities (ablation) | SGP4SR · Clothing | Recall@10: 0.0475 -> 0.0541 | Improved |
| add CGC module (ablation) | SGP4SR · Office | NDCG@10: 0.0862 -> 0.0876 | No clear effect |
| replace image features of items with others (50% mismatch noise) | multimodal sequential recommenders · Baby | Recall@10: SASRecF -23.06%, MissRec -29.48%, HM4SR -11.41%; SGP4SR about -6% | Hurt |
| increase interest center number k beyond 50-60 (ablation) | SGP4SR · Pantry/Baby | Recall@10 | Hurt |
| SGP4SR vs best baseline MissRec | SGP4SR · Pantry | Recall@50: 0.1875 -> 0.1833 | Hurt |
Closest papers by meaning
- Harnessing Multimodal Large Language Models for Multimodal Sequential Recommendation (AAAI 2025 · similarity 0.91)
- MENTOR: Multi-level Self-supervised Learning for Multimodal Recommendation (AAAI 2025 · similarity 0.90)
- Curriculum Conditioned Diffusion for Multimodal Recommendation (AAAI 2025 · similarity 0.89)
- LGMRec: Local and Global Graph Learning for Multimodal Recommendation (AAAI 2024 · similarity 0.88)
- Seeing Beyond Noise: Joint Graph Structure Evaluation and Denoising for Multimodal Recommendation (AAAI 2025 · similarity 0.88)
- MTSTRec: Multimodal Time-Aligned Shared Token Recommender (ICML 2025 · similarity 0.88)
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.
