English

QUACK: Questioning, Understanding, and Auditing Communicated Knowledge in Multimodal Social Deduction Agents

Computation and Language 2026-05-27 v1 Artificial Intelligence Multiagent Systems

Abstract

Social deduction games have become a popular testbed for probing reasoning, deception, coordination, and belief modeling in Large Language Model (LLM) agents. However, most environments are scored only by game outcomes such as win rates and largely remain to text-only interaction, making it difficult to tell whether an agent's language is actually grounded in what it perceived and did, or to identify the failure modes underlying its behavior. To address this gap, we introduce QUACK, an open-source environment and evaluation framework for auditing the grounding of agent language in multimodal social reasoning. QUACK evaluates agents at three levels: game outcomes, behavioral trajectories, and utterance-level consistency. Its core Statement Verification Pipeline reconstructs each agent's ground-truth trajectory from engine logs and checks every discussion claim against it, automatically flagging spatial hallucination, unsupported accusation, deception collapse, and language-action inconsistency. Evaluating three frontier VLMs in both homogeneous and cross-model adversarial settings, we find that even the strongest agent hallucinates 15.1% of its verifiable spatial claims and makes over half of its accusations without grounded evidence. We release the full engine, evaluation framework, toolkit, and logs at https://github.com/AAAAA-Academia-Attractions/QUACK.

Keywords

Cite

@article{arxiv.2605.27068,
  title  = {QUACK: Questioning, Understanding, and Auditing Communicated Knowledge in Multimodal Social Deduction Agents},
  author = {Ye Yuan and Rui Song and Weien Li and Zeyu Li and Haochen Liu and Xiangyu Kong and Changjiang Han and Yonghan Yang and Zichen Zhao and Zixuan Dong and Fuyuan Lyu and Bowei He and Haolun Wu and Jikun Kang and Xue Liu},
  journal= {arXiv preprint arXiv:2605.27068},
  year   = {2026}
}
R2 v1 2026-07-22T07:34:43.186Z