Behavior-Guided Candidate Calibration for Multimodal Recommendation
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917519239938048 |
|---|---|
| author | Li, Zesheng Pan, Chengchang Qi, Honggang |
| author_facet | Li, Zesheng Pan, Chengchang Qi, Honggang |
| contents | Multimodal recommendation benefits from content signals, but the gain depends on how those signals interact with the ranking pipeline. We find that moderate cross-view agreement helps, while stronger agreement suppresses recommendation-specific variation. Spectral analysis shows a clear split: low-frequency components capture shared structure, and higher-frequency components preserve more discriminative signal. Based on this finding, we introduce a behavior-guided candidate calibration model that converts training-only co-user overlap into signed candidate evidence and applies it only to the shortlist produced by the multimodal backbone. The backbone keeps the representation space stable; behavior evidence acts only where ranking is decided. Results on Amazon Baby, Sports, and Electronics show consistent gains over strong multimodal baselines. Code is available at https://github.com/LIZESHENG13/bridge. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_22073 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Behavior-Guided Candidate Calibration for Multimodal Recommendation Li, Zesheng Pan, Chengchang Qi, Honggang Information Retrieval Multimodal recommendation benefits from content signals, but the gain depends on how those signals interact with the ranking pipeline. We find that moderate cross-view agreement helps, while stronger agreement suppresses recommendation-specific variation. Spectral analysis shows a clear split: low-frequency components capture shared structure, and higher-frequency components preserve more discriminative signal. Based on this finding, we introduce a behavior-guided candidate calibration model that converts training-only co-user overlap into signed candidate evidence and applies it only to the shortlist produced by the multimodal backbone. The backbone keeps the representation space stable; behavior evidence acts only where ranking is decided. Results on Amazon Baby, Sports, and Electronics show consistent gains over strong multimodal baselines. Code is available at https://github.com/LIZESHENG13/bridge. |
| title | Behavior-Guided Candidate Calibration for Multimodal Recommendation |
| topic | Information Retrieval |
| url | https://arxiv.org/abs/2605.22073 |