From Individual Excellence to Collective Sustainability: Seeking Strategic Equilibrium in Proactive Multi-Agent Teams

ACL ARR 2026 January Submission9787 Authors

Keywords: Proactive Team Agent, Team Sustainability, Human-AI Collaboration, Nash equilibrium

Abstract: In heterogeneous team settings, proactive AI agents often suffer from Collaborative Myopia: a greedy optimization for immediate task accuracy that ignores long-term team longevity. This leads to the Individual-Centric Trap, where experts (e.g., PIs) are disproportionately overloaded while junior roles remain underutilized, incurring unobserved opportunity costs that erode collective sustainability. To resolve this efficiency-sustainability coupling problem, we propose GT-PMARL (Game-Theoretic Proactive Multi-Agent Reinforcement Learning). We reformulate team coordination as a strategic game that internalizes opportunity cost. Our framework employs: (1) a Positive-Unlabeled scorer to anchor intervention quality under sparse supervision, and (2) a Nash-Pareto competitive objective to seek an equilibrium between individual task excellence and collective load balancing. Empirical experiments on scientific workflows show that GT-PMARL effectively maintains high performance while preventing expert over-exploitation. Our work provides a game-theoretic foundation for building sustainable, balanced human-AI collaborative ecosystems.

Paper Type: Long

Research Area: Human-AI Interaction/Cooperation and Human-Centric NLP

Research Area Keywords: human-AI interaction/cooperation, human-in-the-loop, human-centered evaluation, user-centered design, value-centered design, tool use, function calling, multi-modal agents

Contribution Types: Model analysis & interpretability, NLP engineering experiment, Publicly available software and/or pre-trained models, Data resources

Languages Studied: English

Submission Number: 9787