DC4SR improves sequential recommendation denoising by iteratively calibrating LLM semantic priors and model learning posteriors using their disagreement as a signal for better alignment with true user interests.
In2024 IEEE 40th International Conference on Data Engineering (ICDE)
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
-
Disagreement as Signals: Dual-view Calibration for Sequential Recommendation Denoising
DC4SR improves sequential recommendation denoising by iteratively calibrating LLM semantic priors and model learning posteriors using their disagreement as a signal for better alignment with true user interests.