CETRec improves LLM-based sequential recommendation by adding item-level temporal embeddings and a counterfactual tuning loss that rewards different predictions when temporal order is erased.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Enhancing Temporal Sensitivity of Large Language Model for Recommendation with Counterfactual Tuning
CETRec improves LLM-based sequential recommendation by adding item-level temporal embeddings and a counterfactual tuning loss that rewards different predictions when temporal order is erased.