IFCDSR enriches item representations with frozen CLIP image embeddings and uses separate attention for domain-specific and merged sequences, claiming state-of-the-art results on two Amazon CDSR scenarios.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Image Fusion for Cross-Domain Sequential Recommendation
IFCDSR enriches item representations with frozen CLIP image embeddings and uses separate attention for domain-specific and merged sequences, claiming state-of-the-art results on two Amazon CDSR scenarios.