Mixed-Initiative Context: Structuring and Managing Context for Human-AI Collaboration

2026-04-08

AI Summary

Existing systems often flatten contextual information in multi-turn human-AI interactions, lacking dynamic organization mechanisms and thus becoming susceptible to irrelevant content. This work proposes a “hybrid proactive context” paradigm that reconceptualizes context as an explicit, structured, and actionable object, enabling users and AI to collaboratively and dynamically adjust its structure, scope, and content. Building upon this paradigm, we implement Contextify—a probe system featuring structured context modeling and dynamic management capabilities. User studies demonstrate that our approach significantly enhances users’ sense of control over context, increases their acceptance of AI proactivity, and improves the overall collaborative experience.

Abstract

In the human-AI collaboration area, the context formed naturally through multi-turn interactions is typically flattened into a chronological sequence and treated as a fixed whole in subsequent reasoning, with no mechanism for dynamic organization and management along the collaboration workflow. Yet these contexts differ substantially in lifecycle, structural hierarchy, and relevance. For instance, temporary or abandoned exchanges and parallel topic threads persist in the limited context window, causing interference and even conflict. Meanwhile, users are largely limited to influencing context indirectly through input modifications (e.g., corrections, references, or ignoring), leaving their control neither explicit nor verifiable. To address this, we propose Mixed-Initiative Context, which reconceptualizes the context formed across multi-turn interactions as an explicit, structured, and manipulable interactive object. Under this concept, the structure, scope, and content of context can be dynamically organized and adjusted according to task needs, enabling both humans and AI to actively participate in context construction and regulation. To explore this concept, we implement Contextify as a probe system and conduct a user study examining users' context management behaviors, attitudes toward AI initiative, and overall collaboration experience. We conclude by discussing the implications of this concept for the HCI community.

Problem

• human-AI collaboration
• context management
• mixed-initiative
• interactive context
• multi-turn interaction

Innovation

• Mixed-Initiative Context
• Context Management
• Human-AI Collaboration
• Structured Context
• Interactive Context