Casting top-K retrieval as an MDP over implicit-ALS posteriors with closed-form fold-in transitions, the paper reports that dynamics-aware planning (especially one-step lookahead) outperforms static top-K on multiple datasets under leave-last-n splits when using cosine similarity.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.IR 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
A model-agnostic SID alignment update mitigates staleness from temporal drift in user-item interactions for generative retrievers, improving Recall@K and nDCG@K while reducing compute by 8-9x versus full retraining.
citing papers explorer
-
Planning over Matrix-Factorization MDPs for Candidate Generation
Casting top-K retrieval as an MDP over implicit-ALS posteriors with closed-form fold-in transitions, the paper reports that dynamics-aware planning (especially one-step lookahead) outperforms static top-K on multiple datasets under leave-last-n splits when using cosine similarity.
-
Mitigating Collaborative Semantic ID Staleness in Generative Retrieval
A model-agnostic SID alignment update mitigates staleness from temporal drift in user-item interactions for generative retrievers, improving Recall@K and nDCG@K while reducing compute by 8-9x versus full retraining.