Search papers, labs, and topics across Lattice.
This paper introduces SWIM, a novel list-level evaluator designed to enhance the Generator-Evaluator framework in recommender systems by modeling user behavior as a session-level survival process. By addressing the limitations of traditional evaluators that aggregate point-wise scores and ignore session dynamics, SWIM captures critical contextual dependencies and diminishing returns from repetitive content. Experimental results show that SWIM significantly improves recommendation engagement compared to existing baselines, highlighting its effectiveness in real-world applications like short-video platforms.
SWIM redefines list evaluation by modeling user engagement as a survival process, leading to substantial gains in recommendation effectiveness.
Modern industrial recommender systems have increasingly adopted the Generator-Evaluator (G-E) framework for the re-ranking stage. Within this paradigm, the generator produces candidate item lists from a pool filtered by upstream retrieval and ranking modules, while the evaluator scores these lists and selects the highest-scoring one for final exposure per request. However, on sequential platforms (e.g., short-video apps), users consume items continuously, ignoring artificial list boundaries. Conventional evaluators score lists by aggregating point-wise values, implicitly assuming exposure independence. This fails to capture critical session-level dynamics, such as contextual dependencies, user continuation, and diminishing marginal utility from repetitive content. To bridge this gap, we propose SWIM (Step-Wise Integrated Measure), a list-level evaluator that models user behaviors as a finite-horizon prefix session-level survival process. SWIM estimates the prefix-conditioned contribution of the current list to the session-level objective by factorizing it into a recursive survival distribution and reached-position conditional rewards. Leveraging a causally-masked Transformer, SWIM efficiently estimates continuation probabilities and utilities in parallel, satisfying strict industrial latency constraints. Extensive experiments demonstrate that SWIM significantly outperforms baselines in listwise reranking tasks, yielding substantial improvements in overall recommendation engagement.
