Talk is Cheap, Communication is Hard: Dynamic Grounding Failures and Repair in Multi-Agent Negotiation

TL;DR

This paper investigates the challenges of dynamic grounding in multi-agent negotiation, revealing common failure modes and emphasizing the importance of interaction in establishing mutual understanding.

Contribution

It introduces a multi-turn negotiation game to analyze grounding failures and identifies key factors hindering successful coordination among agents.

Findings

Agents fail to reach Pareto-optimal allocations in multi-turn negotiations.

Four failure modes are identified: history loss, anchoring, defaulting to equal splits, and referential errors.

Coordination gaps are due to dynamic grounding issues, not individual reasoning limits.

Abstract

Grounding is the collaborative process of establishing mutual belief sufficient for a communicative goal. While static grounding maps language to a shared context, dynamic grounding requires agents to negotiate meaning across turns. Current multi-agent Large Language Model (LLM) benchmarks largely emphasize static, one-shot tasks, overlooking whether agents can repair grounding breakdowns through interaction. We introduce an iterated multi-turn negotiation game where two agents allocate shared resources to private projects with verifiable jointly optimal outcomes. Although individual agents can identify Pareto-optimal allocations in isolation, agent dyads consistently fail to reach them across models. We identify four failure modes: (1) loss of shared interaction history, (2) stubborn anchoring to early proposals, (3) defaulting to equal splits over reward-maximizing coordination, and (4) referential binding errors across turns. Our baselines show that the coordination gap is not explained by individual reasoning limits or insufficient information exchange alone. Instead, the bottleneck lies in dynamic grounding: joint plan formation, commitment, and execution.