Modeling Distinct Human Interaction in Web Agents
This work addresses the need for more adaptive and collaborative web agents by improving human-agent interaction, though it is incremental as it builds on existing agentic systems with a focus on intervention modeling.
The paper tackled the problem of autonomous web agents lacking understanding of human intervention patterns, by modeling distinct human interaction styles to predict when users intervene, resulting in a 61.4-63.4% improvement in prediction accuracy and a 26.5% increase in user-rated agent usefulness.
Despite rapid progress in autonomous web agents, human involvement remains essential for shaping preferences and correcting agent behavior as tasks unfold. However, current agentic systems lack a principled understanding of when and why humans intervene, often proceeding autonomously past critical decision points or requesting unnecessary confirmation. In this work, we introduce the task of modeling human intervention to support collaborative web task execution. We collect CowCorpus, a dataset of 400 real-user web navigation trajectories containing over 4,200 interleaved human and agent actions. We identify four distinct patterns of user interaction with agents -- hands-off supervision, hands-on oversight, collaborative task-solving, and full user takeover. Leveraging these insights, we train language models (LMs) to anticipate when users are likely to intervene based on their interaction styles, yielding a 61.4-63.4% improvement in intervention prediction accuracy over base LMs. Finally, we deploy these intervention-aware models in live web navigation agents and evaluate them in a user study, finding a 26.5% increase in user-rated agent usefulness. Together, our results show structured modeling of human intervention leads to more adaptive, collaborative agents.