Learning Self-Modulating Attention in Continuous Time Space with
          Applications to Sequential Recommendation

Chao Chen 1 Haoyu Geng 1 2 Nianzu Yang 1 2 Junchi Yan 1 2 Daiyue Xue 3 Jianping Yu 3 Xiaokang Yang 1 2

              Abstract                  fade away due to maturity or a change in lifestyle. Nonethe-
                                   less, it is difficult to model all causal aspects explicitly
   User interests are usually dynamic in the real
                               ...