HOW TO MODEL HUMAN ACTIONS DISTRIBUTION WITH EVENT SEQUENCE DATA

ICLR 2026 Conference Submission20176 Authors

Keywords: event sequences

Abstract: This paper studies forecasting of the future distribution of events in human action sequences, a task essential in domains like retail, finance, healthcare, and recommendation systems where the precise temporal order is often less critical than the set of outcomes. We challenge the dominant autoregressive paradigm and investigate whether explicitly modeling the future distribution or order-invariant multi-token approaches outperform order-preserving methods. We analyze local order invariance and introduce a KL-based metric to quantify temporal drift. We find that a simple explicit distribution forecasting objective consistently surpasses complex implicit baselines. We further analyze the emergence of mode collapse in predicted categories, identifying and evaluating key contributing mechanisms. This work provides a principled framework for selecting modeling strategies and offers practical guidance for building more accurate and robust forecasting systems.

Primary Area: learning on time series and dynamical systems

Submission Number: 20176