ProACT: Towards Breakdown-Aware Proactive Agent in Multi-User Collaboration

2026-07-04

AI Summary

This work addresses the limitation of existing conversational agents in multi-party collaboration, which typically respond passively and struggle to proactively identify and mitigate communication breakdowns such as ambiguous goals, circular discussions, or imbalanced participation. To bridge this gap, the authors propose ProACT, a novel proactive agent framework grounded in theories of common ground, collaborative planning, and coordinated action. ProACT analyzes speaker-attributed dialogue history to detect collaboration barriers in real time and dynamically invokes targeted intervention skills. As the first proactive agent framework designed specifically for multi-party settings, ProACT introduces a dedicated benchmark comprising 3,244 dialogue turns across six collaborative scenarios. Experimental results demonstrate that ProACT significantly outperforms direct chat baselines in intervention appropriateness, non-intrusiveness, conciseness, and overall quality, marking a pivotal shift from reactive response to active collaboration.

Abstract

Conversational agents are increasingly embedded in human collaborative work, yet they remain fundamentally passive and reactive: they respond to explicit user requests rather than proactively recognizing moments when a team would benefit from timely intervention as human collaborators often do. This reactive design substantially limits the use of agents as active participants in multi-user collaboration, where disagreements, ambiguous goals, forgotten constraints, underspecified plans, discussion loops, and imbalanced participation can gradually undermine group progress. To move agents from passive assistants toward active participants in multi-user collaboration, we introduce ProACT, a breakdown-aware agent framework grounded in theories of common ground, collaborative planning, and coordination work. ProACT observes the speaker-attributed conversation history, determines whether the current turn contains a collaboration breakdown requiring intervention, decides whether the agent should stay silent or speak, and, when speaking is needed, routes the case to a targeted collaboration skill. We further introduce the first multi-user collaboration benchmark for evaluating proactive agents across project planning, product design, research collaboration, logistics, education, and resource-constrained decision making. Across 3,244 turn-level examples and five LLM backbones, ProACT consistently improves collaborative appropriateness, non-interruptiveness, conciseness, and judged intervention quality over direct chat.

Problem

• proactive agent
• multi-user collaboration
• collaboration breakdown
• conversational AI
• common ground

Innovation

• proactive agent
• collaboration breakdown
• multi-user collaboration
• common ground
• intervention routing