Gentiane Venture
Professor of robotics, the University of Tokyo
About the Workshop
Objective
Explore how multimodal data can be systematically transformed into adaptive robot behaviors for next-generation Human–Robot Interaction (HRI). We focus on the end-to-end pipeline from clean and generalizable data curation, through learning and modeling, to real-time adaptive behavior—with emphasis on human internal states such as mind wandering, anxiety, engagement, and trust.
Expected Outcomes
Synthesize best practices, identify open research questions, and foster cross-disciplinary collaborations. A post-workshop summary will be published, and follow-up submissions encouraged to relevant venues.
Call for Participation
We welcome short papers, extended abstracts, and posters that advance human state-aware robotics for real-world HRI.
What to submit
Submit work that connects multimodal human data to adaptive robot behavior—methods, datasets, studies, systems, or lessons learned. We particularly encourage contributions that address robustness, responsible data practices, and deployment in-the-wild.
- Submission format: Either extended abstracts (2–4 pages, excluding references) or full papers (6 pages, excluding references), using the IEEE paper format.
- Templates: LaTeX template, MS Word template, LaTeX template on Overleaf.
- Submission link: Submit your paper using this link
Accepted submissions will be archived on the workshop’s website.
Poster visibility: Poster presenters prepare a 1-minute lightning talk video. Videos are promoted on social media, played during breaks, and linked on the workshop website.
🏆 The workshop will recognize outstanding poster presentations by awarding first, second, and third place prizes.
Multimodal Sensing
Gaze, speech, motion, physiological signals for HRI.
Human State Estimation
Engagement, stress, fatigue, mind wandering detection.
Data Challenges
Alignment, missing data, dataset bias, robustness.
Adaptive Learning
Supervised, RL, imitation, LLM-based interaction policies.
Online Adaptation
Real-time learning and human-in-the-loop systems.
Trustworthy Robotics
Calibration, over-reliance, transparency, human agency.
Ethics & Privacy
Biosignals in the wild: privacy, consent, data governance.
Applications
Social robots, teleoperation, assistive robots, safety-critical HRI.
Responsible Sharing
De-identification, access control, reproducibility best practices.
Invited Speakers
Leading experts sharing complementary perspectives.
Workshop Program
| Time | Activity |
|---|---|
| 8:45 | Welcome and Introduction |
| 9:00 | Invited Talk 1 — Prof. Tetsuya Ogata |
| 9:20 | Invited Talk 2 — Prof. Gentiane Venture |
| 9:40 | Breakout Group Discussion |
| 10:00 | Coffee Break + Posters + Lightning Talk Videos |
| 10:20 | Invited Talk 3 — Prof. Kristiina Jokinen |
| 10:40 | Invited Talk 4 — Prof. Silvia Rossi |
| 11:00 | Breakout Group Discussion |
| 11:20 | Coffee Break + Posters + Lightning Talk Videos |
| 11:40 | Panel Discussion & Lessons-Learned Report |
| 12:10 | Closing |
Organizing Committee
An interdisciplinary team spanning HRI, AI, and robotics.
Intended Audience
Who should attend this workshop?
This workshop targets an interdisciplinary audience of researchers and practitioners in Human–Robot Interaction (HRI), robotics, artificial intelligence, and related fields who are interested in transforming multimodal human data into adaptive robot behavior. It is particularly relevant for those working on human state understanding, multimodal sensing and fusion, learning and modeling for interaction, and real-time behavioral adaptation.
The workshop will also attract researchers concerned with trustworthy and responsible robotics, including issues of human agency, over-reliance, and ethical use of sensitive data, as well as those developing or deploying adaptive robotic systems in domains such as social interaction, teleoperation, and safety-critical applications.
Human State Understanding
Modeling internal states during interaction.
Multimodal Sensing & Fusion
Combining signals reliably across modalities.
Learning & Modeling for Interaction
Policies and models grounded in HRI.
Real-time Behavioral Adaptation
Online inference and action selection.
Trustworthy & Responsible Robotics
Agency, calibration, transparency.
Safety-Critical Applications
Robustness for real-world deployment.
Sponsors
This workshop is sponsored by :
Contact
For questions, sponsorship, or participation inquiries, reach out to the workshop team.
For inquiries about the workshop, please contact the organizing committee at: