QUACK: Questioning, Understanding, and Auditing Communicated Knowledge in Multimodal Social Deduction Agents
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917536495304704 |
|---|---|
| author | Yuan, Ye Song, Rui Li, Weien Li, Zeyu Liu, Haochen Kong, Xiangyu Han, Changjiang Yang, Yonghan Zhao, Zichen Dong, Zixuan Lyu, Fuyuan He, Bowei Wu, Haolun Kang, Jikun Liu, Xue |
| author_facet | Yuan, Ye Song, Rui Li, Weien Li, Zeyu Liu, Haochen Kong, Xiangyu Han, Changjiang Yang, Yonghan Zhao, Zichen Dong, Zixuan Lyu, Fuyuan He, Bowei Wu, Haolun Kang, Jikun Liu, Xue |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_27068 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | QUACK: Questioning, Understanding, and Auditing Communicated Knowledge in Multimodal Social Deduction Agents Yuan, Ye Song, Rui Li, Weien Li, Zeyu Liu, Haochen Kong, Xiangyu Han, Changjiang Yang, Yonghan Zhao, Zichen Dong, Zixuan Lyu, Fuyuan He, Bowei Wu, Haolun Kang, Jikun Liu, Xue Computation and Language Artificial Intelligence Multiagent Systems 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. |
| title | QUACK: Questioning, Understanding, and Auditing Communicated Knowledge in Multimodal Social Deduction Agents |
| topic | Computation and Language Artificial Intelligence Multiagent Systems |
| url | https://arxiv.org/abs/2605.27068 |