Interaction, Process, Infrastructure: A Unified Framework for Human-Agent Collaboration

TL;DR

This paper introduces a layered framework for human-agent collaboration that emphasizes process management and structural adaptation, aiming to create more cohesive and dynamic AI-assisted work systems.

Contribution

It presents a novel layered framework integrating Interaction, Process, and Infrastructure, with a focus on explicit process representation and dynamic reorganization for improved collaboration.

Findings

Proposes a five-module Process Model for structural adaptation.

Defines Structural Adaptation as a key to dynamic collaboration.

Provides a unified theoretical foundation for human-agent systems.

Abstract

While AI tools are increasingly prevalent in knowledge work, they remain fragmented, lacking the architectural foundation for sustained, adaptive collaboration. We argue this limitation stems from their inability to represent and manage the structure of collaborative work. To bridge this gap, we propose a layered conceptual framework for human-agent systems that integrates Interaction, Process, and Infrastructure. Crucially, our framework elevates Process to a first-class concern, an explicit, inspectable structural representation of activities. The central theoretical construct is Structural Adaptation, enabling the process to dynamically reorganize itself in response to evolving goals. We introduce a five-module Process Model as the representational basis for this adaptation. This model offers a unified theoretical grounding, reimagining human-agent collaboration as a coherent system for complex real-world work.