MVCrec improves next-item prediction by running three contrastive losses (within-ID, within-graph, and cross-view) and fusing the resulting representations with global-plus-local attention.
Im- proving contrastive learning with model augmentation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ID and Graph View Contrastive Learning with Multi-View Attention Fusion for Sequential Recommendation
MVCrec improves next-item prediction by running three contrastive losses (within-ID, within-graph, and cross-view) and fusing the resulting representations with global-plus-local attention.