Program
The workshop takes place on Friday, October 9, 2026 in Naples, Italy, as part of ICMI 2026.
| Time | Session | Authors / Presenters | Format |
|---|---|---|---|
| 9:00 – 9:10 | Intro | Program | |
| 9:10 – 10:00 | Keynote: Can We Trust What We Generate? Evaluating Generative Multimodal Systems in Immersive Health Applications | Paola Barra | Program |
| 10:00 – 10:30 | Official Break | Break | |
| 10:30 – 11:00 | Multimodal Depression Detection Varies with Cross-Dataset Differences | Maneesh Bilalpur; Saurabh Hinduja; Sonish Sivarajkumar; Nicholas Allen; Yanshan Wang; Itir Onal Ertugrul; Jeff Cohn | Talk |
| 11:00 – 11:30 | Scenario-Adaptive Multimodal Emotion Recognition for Supporting Affect Perception in Doctor–Patient Consultations | Zhaolong Wang; Aoi Taguchi; Homura Kawamura; Tomohumi Miura; Yuka Maeda; Yukihiko Okada; Keiichi Zempo | Talk |
| 11:30 – 12:00 | Retrieval-Augmented Personalization with Foundation Models for Wearable Stress Detection | Louis Simon; Mohamed Chetouani | Talk |
| 12:00 – 12:30 | Buffer / Free Time / Poster Preparation | Break | |
| 12:30 – 14:00 | Lunch | Break | |
| 14:00 – 14:02 | PlainHealth: A Multimodal Service Suite for Cognitively Accessible Healthcare Communication | Lourdes Moreno; Paloma Martínez; Jesus M. Sanchez-Gomez | Poster pitch |
| 14:02 – 14:04 | Benchmarking Vector Quantized Auto-Encoders for Multimodal Vision-Language Tokenization of Medical Images | Emílio Dolgener Cantú; Manasi Acharya; Jim Berend; Md Golam Rasul; Jackie Ma | Poster pitch |
| 14:04 – 14:06 | Co-Designing Accessible Embodied Conversational Agents for Post-COVID Syndrome Patients: A Participatory Case Study | Roman de Laporte; Nicolas Kaulen; Ana Kirschbaum; Chantal Herrmann; Caterina Neef; Anja Richert | Poster pitch |
| 14:06 – 14:08 | AI-based Analysis and Generation of Nonverbal Behavior: NEUROGES® as a Neuropsychologically Grounded Behavior Taxonomy for Health Applications | Hedda Lausberg; Philipp Müller; Patrick Gebhard | Poster pitch |
| 14:08 – 14:10 | A Modular Reference Architecture for Patient-Facing Generative AI Health Coaches in Chronic Self-Management | Leon Paul Mondrian Munz; Ana Kirschbaum; Christopher Kania; Tim Fellerhoff; Anja Richert | Poster pitch |
| 14:10 – 15:30 | Poster Session (Discussion) | Poster session | |
| 15:30 – 16:00 | Official Break | Break | |
| 16:00 – 16:30 | Wrap Up & Next Steps | Program |
Keynote
Can We Trust What We Generate? Evaluating Generative Multimodal Systems in Immersive Health Applications
Paola Barra — University of Naples Parthenope, Italy
Generative and multimodal AI are entering immersive environments, from the virtual spaces already used in therapy, such as exposure therapy, to conversational agents that respond to facial expressions, gaze and physiological signals. When health is involved, the question shifts from “does it work?” to “how do we know it works, and why should anyone trust it?”
Drawing on our recent work on integrating generative AI and virtual reality in mental health interventions, including a user study of a proposed co-creation framework, and on a user study of how trust changes when generative models produce false information, this talk discusses two open problems. The first is evaluation: what we can actually claim when studies show feasibility and acceptability but not yet clinical efficacy, and why current measures across the field remain small-scale and inconsistent. The second is trust: hallucinations in immersive settings, user control and co-design, transparency, and the handling of biometric data.
The talk closes with open questions for the HCI, ML and clinical communities, to open the discussion for the rest of the day.
About the speaker
Paola Barra is a tenure-track Assistant Professor (RTD-B) in Computer Science at the Department of Science and Technology, University of Naples Parthenope, and holds the Italian national qualification as Associate Professor. She received her PhD in Computer Science from the University of Salerno in 2021. Her research covers computer vision, biometrics, affective computing and multimodal interaction, with a recent focus on generative AI in virtual and extended reality for mental health, and on how such systems can be evaluated and trusted. She has served as Guest Editor for several special issues and as Program Chair of CHItaly 2025, and she co-founded the university spin-off Commigo S.r.l.