LDMI reports consistent accuracy gains over state-of-the-art multi-interest recommenders on Amazon Beauty, Books, and Video Games by combining LLM-driven semantic clusters with collaborative capsules and crowd-level contrastive learning.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
LLM-Driven Dual-Level Multi-Interest Modeling for Recommendation
LDMI reports consistent accuracy gains over state-of-the-art multi-interest recommenders on Amazon Beauty, Books, and Video Games by combining LLM-driven semantic clusters with collaborative capsules and crowd-level contrastive learning.