Learning Next Action Predictors from Human-Computer Interaction
Authors: Omar Shaikh, Valentin Teutschbein, Kanishk Gandhi et al.
Date: 2026-04-21
Summary
Learns to predict user’s next actions from observing human-computer interaction patterns. Relevant to proactive AI agents that anticipate user intent.
Concepts
- Agent Memory — learning from interaction history
- Context Engineering
See Also
- generative-user-models — predictive user modeling
- tweet-curiosity-reward-personalization — curiosity-driven personalization