This ICMI 2026 workshop aims to establish an international research platform to explore how linguistic and cultural differences shape nonverbal behavior and interaction dynamics. Building on the success of our first workshop, this second edition shifts the focus from problem identification to concrete action and methodological evaluation.
This year’s discussion-centric workshop focuses on two key themes:
Concrete Data Collection & Case Studies: Sharing practical "Case Reports" to overcome the logistical hurdles of multi-site data collection, with the goal of drafting a roadmap for truly cross-cultural multimodal datasets.
Evaluating MLLMs in Cultural Contexts: Developing methodologies to benchmark Multimodal Large Language Models (MLLMs) for cultural sensitivity, specifically examining their ability to handle cultural nuances in gestures, facial expressions, and turn-taking.
We have accepted 6 papers (3 long, 3short) (23rd, July)
The keynote talk has been confirmed (26th, May 2026)
The workshop has been accepted for ICMI 2026! See you in Napoli🙋 (20th, February 2026)
It includes (but is not limited to) the following:
Cross-cultural differences in nonverbal communication
Data collection methodologies for intercultural interaction
Cultural variations in gestures, facial expressions, back-channeling, and turn-taking
Multimodal machine learning and signal processing methods for detecting cultural differences
Cross-linguistic differences and universals in turn-taking behavior
Paper Submission July 12th, 2026 (extended)
Paper Notification July 24th, 2026
Camera Ready July 31st, 2026
Workshop Day October 9th, 2026 (Afternoon)
We invite both long papers (up to 8 pages) and short papers (up to 4 pages), all formatted in the double-column ACM conference style used by ICMI 2026. Pages consisting solely of references do not count toward the page limit for either paper type. All submissions must be prepared for double-blind review and submitted as PDF files via PCS.
Submission Guidelines for Authors (ICMI 2026):
https://icmi.acm.org/2026/guidelines/
Submission Site:
https://new.precisionconference.com/submissions/icmi26a
You will need an PCS account, and we strongly recommend creating one as soon as possible if you haven't already.
Elisabeth André (Augsburg University)
Abstract
While certain social signals can be observed across cultures, cultures differ in when, whether, and how such behaviors are displayed. Developing culturally-aware conversational agents, therefore, requires interpreting and generating behavior in relation to culture-specific display rules and social norms. In this talk, I first report on research on cross-cultural multimodal interaction datasets, such as NoXi+J, and on analyses of social behavior conducted in the MultiMediate challenge series, which investigate engagement and conversational behaviors across participants from European and Asian cultural backgrounds. Building on these analyses, the second part of the talk addresses the generation of culturally aware behaviors for socially interactive agents. I discuss our approach for synthesizing culturally adaptive listener behaviors that integrates cultural identity as a conditional factor into a real-time generative model. The talk concludes with a discussion of open challenges for culturally aware AI, including modeling culture-specific behaviors without reinforcing stereotypes.
Biography
Elisabeth André is a Full Professor of Computer Science and the Founding Chair of Human-Centered Artificial Intelligence at Augsburg University, Germany. Her research has been foundational in enabling machines to perceive and respond to human multimodal conversational behaviors in a natural and socially intelligent manner. For her pioneering contributions to socially-interactive artificial intelligence, she was awarded the ICMI Sustained Accomplishment Award and the Gottfried Wilhelm Leibniz Prize by the German Research Foundation (DFG).
14:00-14:10 Opening
14:10-15:00 Invited Talk
15:05-15:30 Oral Presentations (1)
Judging Turn-Taking Through a Monocultural Lens: A Parallel-Corpus Probe of Cultural Bias in LLMs
(Nur Keleşoğlu) - Long
Enabling AI to Find Multimodal Signals of Storytelling Cross-Culturally: A Multi-Step Approach
(Anna Bonifazi, Juman Al-qaoud, Daniel Alcaraz Carrión, Giorgia Crosilla, Sandra Debreslioska, Maria Del Rosario Illan Castillo, Øyvind Eide, Madeleine Frings, Armine Garibyan, Sonja Gipper, Martine Grice, Yassine Iabdounane, Cristobal Pagan Canovas, Malin Patel, Felix Rau, Raul Sanchez, Mariem Somrani Achouri, Mark Turner, Peter Uhrig, Fabio Vitali, Mengyuan Yu) - Short
15:30-16:00 Coffee Break
16:00-16:50 Oral Presentations (2)
Toward a Culturally Adapted Chinese Language Agent: A Wizard-of-Oz Study of Nonverbal Behavior in Chinese-German Intercultural Interaction
(Siddhant Jain, Anna Lea Reinwarth, Habil. Dimitra Tsovaltzi, Rafael Math, Julia Renner) - Long
Browser-Based Deployment of Spoken Multi-Agent Experiments for Multi-Site Research
(Leo Huovinen, Kristiina Jokinen, Taiga Mori, Markku Turunen) - Short
TEIDAN: A Multilingual Multiparty Dialogue Corpus
(Taiga Mori, Koji Inoue, Mikey Elmers, Divesh Lala, Tatsuya Kawahara) - Long
Evaluating Cross-Cultural Generative Embodied Social Agents
(Marius Funk, Shogo Okada, Elisabeth André) - Short
16:55-17:50 Group Discussion
17:50-18:00 Closing (Summary)
Note that Long and Short papers are assigned 15 min. and 10 min. (including Q&A), respectively.
The main contact address of the workshop is: ccmi-organizer@googlegroups.com
Fu Yahui (Nagoya University, Japan)
Zhaojie Luo (Southeast University, Nanjing, China)
Li Sixia (Japan Advanced Institute of Science and Technology, Japan)
Ryo Ishii (NTT, Japan)
Hung-Hsuan HUANG (University of Fukuchiyama, Japan)
Shun Katada (Wakayama University, Japan)
Hiroki Tanaka (International Christian University, Japan)
Candy Olivia Mawalim (Japan Advanced Institute of Science and Technology, Japan)
Fuminori Nagasawa (Nagoya Institute of Technology, Japan)