🤖 AI Summary
This study addresses coordination inefficiencies in human–AI collaboration within shared workspaces, where the absence of effective coordination mechanisms often incurs process losses and can even reduce team performance when new collaborators are introduced. To mitigate these issues, this work proposes a scaffolding mechanism that integrates shared group memory with human-in-the-loop (HITL) approval gating, employing structured coordination strategies to optimize responsibility allocation and expert knowledge scheduling. Evaluated across 1,482 experimental sessions using the Collaborative Gym environment and DiscoveryBench tasks, the approach significantly enhances joint decision-making performance in three-person teams, sharpens the clarity of responsibility signaling, and more effectively channels expert knowledge to guide collective action.
📝 Abstract
Automated AI agents are increasingly capable, yet many scientific and professional tasks require human judgment and contextual expertise. We study shared-workspace human-AI teams, where AI agents and human collaborators must coordinate responsibilities before submitting a final answer. Using the Collaborative Gym environment with DiscoveryBench tasks, we examine when adding simulated human collaborators improves performance and when process loss turns additional collaborators into coordination overhead. Across 1,482 sessions, adding relevant collaborators can lower performance when teams lack structure to coordinate their contributions. We then evaluate scaffolding that combines shared group memory with simulated human-in-the-loop (HITL) gates, where selected actions require approval from a designated simulated participant. This scaffolding yields higher mean performance, most clearly in three-person teams, with clearer responsibility signals and stronger routing of expertise to team actions. Overall, how human-AI teams coordinate and integrate expertise matters as much as the capability available to them.