A half-day workshop on social, embodied, and multimodal human-AI interaction
This workshop focuses on social and embodied intelligence for multimodal human-AI interaction. It brings together research on multimodal perception, affective computing, speech and vision, embodied AI, and human-centered interaction design.
We are interested in AI systems that can perceive, interpret, and generate socially relevant multimodal behavior, including speech, gestures, facial expressions, body movement, and contextual cues. The goal is to better understand how such systems can support more natural, adaptive, and meaningful interaction with people.
The workshop provides a forum for sharing recent advances, discussing open challenges, and connecting researchers working across AI, HCI, computer vision, speech processing, and affective computing.
Tentative Program
- 10 minWelcome and introduction
- 30 minInvited talk 1
- 60 minAccepted paper presentations
- 30 minInvited talk 2
- 15 minPanel / open discussion
- 10 minConclusion & Closing
Workshop Topics
The workshop welcomes contributions on social and embodied intelligence for human-AI interaction, including research on speech, video, text, behavioral cues, physiological and interaction signals, with topics including but not limited to:
Emotion and Social Intelligence
- Emotion recognition, affective computing, and social intelligence from multimodal, physiological, or interaction signals
- Emotion-aware, socially adaptive, and context-sensitive human-AI interaction
- Social signal processing, engagement analysis, rapport, empathy, and conversational dynamics
- Modeling user states, intentions, and interaction outcomes in real-world settings
Multimodal Perception and Understanding
- Speech, audio, video, text, and sensor-based perception for human behavior understanding
- Signal processing, feature learning, representation learning, and foundation models for human-centered AI
- Human behavior analysis, social signal analysis, and temporal modeling in real-world interaction data
- Robust, efficient, and data-aware methods for understanding complex social and embodied behavior
Multimodal Generation and Embodied Interaction
- Generative models for interactive AI, conversational systems, and adaptive user experiences
- Talking avatars, digital humans, embodied conversational agents, and socially aware robotics
- Interaction modeling for hybrid human-AI systems across language, action, and perception
- Applications to healthcare, education, accessibility, wellbeing, and collaborative work
Call for Presentations
We invite participants to contribute to the workshop by submitting either an abstract or a workshop paper to be included in the collected volume of the HHAI 2026 Workshop and Tutorial Proceedings under CEUR-WS.
Participants can choose from the following submission types:
- Abstract (250 words): A concise summary of your research focus or a selected overview of relevant studies in hybrid intelligence.
- Short Paper (5-9 pages): Work-in-progress findings or pilot study results that contribute to the evolving research landscape.
- Full Research Paper (10-12 pages): Comprehensive study results, including theoretical advances, empirical findings, or methodological innovations.
All submissions must follow the CEUR template.
- LaTeX template: Access on Overleaf
The workshop paper should be submitted here: https://cmt3.research.microsoft.com/SEIHHAI2026
Important Dates
March 14, 2026: Call for Submissions
May 15, 2026
May 25, 2026:
Submission Deadline for Research Abstract / Papers
May 29, 2026: Notifications of Acceptance
June 5, 2026: Camera-Ready Deadline for Workshop Papers
July 7, 2026: Workshop Day 2 (afternoon), HHAI 2026 Workshops, Brussels
The Venue
Brussels, Belgium
Contact
For workshop enquiries: fang.kang@oulu.fi, chen.haoyu@oulu.fi, yueyi.yang@oulu.fi
Acknowledgment
The Microsoft CMT service was used for managing the peer-reviewing process for this conference. This service was provided for free by Microsoft and they bore all expenses, including costs for Azure cloud services as well as for software development and support.