作者位次 #3/5SIGIR 2026Poster· Information Retrieval (Short)
Revisiting the Role of Learned Attention Weighting in SASRec
Keito Kozaki · Keigo Sakurai · Ren Togo · Takahiro Ogawa · Miki Haseyama
摘要
Causal self-attention models such as SASRec are widely used in sequential recommendation, where learned attention weights are often assumed to provide crucial importance weighting over past interactions. Yet it is unclear when predictive performance truly depends on such non-uniform weighting. We...