ACL Findings2023
Werewolf Among Us: Multimodal Resources for Modeling Persuasion Behaviors in Social Deduction Games
Bolin Lai, Hongxin Zhang, Miao Liu, Aryan J. Pariani, Fiona Ryan, Wenqi Jia, Shirley Anugrah Hayati, James M. Rehg, Diyi Yang
Abstract
Persuasion modeling is a key building block for conversational agents. Existing works in this direction are limited to analyzing textual dialogue corpora. We argue that visual signals also play an important role in understanding human persuasive behaviors. In this paper, we introduce the first multimodal dataset for modeling persuasion behaviors. Our dataset includes 199 dialogue transcriptions and videos captured in a multi-player social deduction game setting, 26, 647 utterance level annotations of persuasion strategy, and game level annotations of deduction game outcomes. We provide extensive experiments to show how dialogue context and visual signals benefit persuasion strategy prediction. We also explore the generalization ability of language models for persuasion modeling and the role of persuasion strategies in predicting social deduction game outcomes. Our dataset can be found at https://persuasion-deductiongame. socialai-data.org. The codes and models are available at https://github.com/ SALT-NLP/PersuationGames.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
8 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- Omni-MMSI: Toward Identity-attributed Social Interaction Understanding cites Ego4D: Around the World in 3, 000 Hours of Egocentric Video
- Omni-MMSI: Toward Identity-attributed Social Interaction Understanding cites Multi-speaker Attention Alignment for Multimodal Social Interaction
- Ego4D: Around the World in 3, 000 Hours of Egocentric Video cites Multiscale Vision Transformers
- Ego4D: Around the World in 3, 000 Hours of Egocentric Video cites Language Models are Few-Shot Learners
- Multiscale Vision Transformers cites Language Models are Few-Shot Learners
- Multi-speaker Attention Alignment for Multimodal Social Interaction cites Ego4D: Around the World in 3, 000 Hours of Egocentric Video
- Multi-speaker Attention Alignment for Multimodal Social Interaction cites MultiMind: Enhancing Werewolf Agents with Multimodal Reasoning and Theory of Mind
- VideoCLIP: Contrastive Pre-training for Zero-shot Video-Text Understanding cites HERO: Hierarchical Encoder for Video+Language Omni-representation Pre-training
- VideoCLIP: Contrastive Pre-training for Zero-shot Video-Text Understanding cites Language Models are Few-Shot Learners
Cited by top-tier papers8
- Multi-speaker Attention Alignment for Multimodal Social InteractionLiangyang Ouyang, Yifei Huang, Mingfang Zhang, Caixin Kang et al.CVPR 2026 · 8 citations
- Bayesian Social Deduction with Graph-Informed Language ModelsShahab Rahimirad, Guven Gergerli, Lucia Romero, Angela Qian et al.ACL 2026 · 4 citations
- Omni-MMSI: Toward Identity-attributed Social Interaction UnderstandingXinpeng Li, Bolin Lai, Hardy Chen, Shijian Deng et al.CVPR 2026 · 3 citations
- MultiMind: Enhancing Werewolf Agents with Multimodal Reasoning and Theory of MindZheng Zhang, Nuoqian Xiao, Qi Chai, Deheng Ye et al.ACM MM 2025 · 1 citation
- The Stackelberg Speaker: Optimizing Persuasive Communication in Social Deduction GamesZheng Zhang, Deheng Ye, Peilin Zhao, Hao WangACL 2026
Builds on8
- Language Models are Few-Shot LearnersTom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah et al.NeurIPS 2020 · 64,255 citations
- Multiscale Vision TransformersHaoqi Fan, Bo Xiong, Karttikeya Mangalam, Yanghao Li et al.ICCV 2021 · 1,611 citations
- Ego4D: Around the World in 3, 000 Hours of Egocentric VideoKristen Grauman, Andrew Westbury, Eugene Byrne, Zachary Chavis et al.CVPR 2022 · 525 citations
- VideoCLIP: Contrastive Pre-training for Zero-shot Video-Text UnderstandingHu Xu, Gargi Ghosh, Po-Yao Huang, Dmytro Okhonko et al.EMNLP 2021 · 399 citations
- HERO: Hierarchical Encoder for Video+Language Omni-representation Pre-trainingLinjie Li, Yen-Chun Chen, Yu Cheng, Zhe Gan et al.EMNLP 2020 · 387 citations
Related papers
- Persuasion Strategies in AdvertisementsYaman Kumar, Rajat Jha, Arunim Gupta, Milan Aggarwal et al.AAAI 2023
- Long-Horizon Dialogue Understanding for Role Identification in the Game of Avalon with Large Language ModelsSimon Stepputtis, Joseph Campbell, Yaqi Xie, Zhengyang Qi et al.EMNLP Findings 2023 · 2 citations
- SAD: A Large-Scale Strategic Argumentative Dialogue DatasetYongkang Liu, Jiayang Yu, Mingyang Wang, Yiqun Zhang et al.ACL 2026
- Persuading across Diverse Domains: a Dataset and Persuasion Large Language ModelChuhao Jin, Kening Ren, Lingzhen Kong, Xiting Wang et al.ACL 2024 · 9 citations
- Selective Vision is the Challenge for Visual Reasoning: A Benchmark for Visual Argument UnderstandingJiwan Chung, Sungjae Lee, Minseo Kim, Seungju Han et al.EMNLP 2024 · 4 citations
