A plug-in contrastive loss that enforces monotonicity between numerical features and recommender outputs, via counterfactual sample synthesis, improves AUC, GAUC, and monotonicity.
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
-
Enhancing Interpretability and Effectiveness in Recommendation with Numerical Features via Learning to Contrast the Counterfactual samples
A plug-in contrastive loss that enforces monotonicity between numerical features and recommender outputs, via counterfactual sample synthesis, improves AUC, GAUC, and monotonicity.