SCOPE
Socially adaptive robots need to interpret human communication and respond appropriately as users, tasks, and environments change. Research in social signal processing and multimodal machine learning provides foundations for combining speech, gaze, facial expressions, gestures, and other contextual information to understand human behaviour. These capabilities support a central question for HRI: how can robots translate multimodal observations into meaningful and appropriate social responses? [1]
This workshop focuses on the connection between social understanding and adaptive robot behaviour. Building on research linking user modelling to behavioural adaptation, it explores how robots can represent human states, intentions, and preferences, reason about situations, and personalise their interactions over time [2]. The workshop explores how multimodal fusion and grounding, social reasoning, and cognitive architectures can enable robots to learn continuously and adapt through human feedback.
The workshop also aims to address the responsible development and evaluation of adaptive robot behaviour. Evidence linking robot performance and characteristics to human trust highlights the importance of considering users’ perceptions alongside technical capabilities [3]. The workshop therefore welcomes methods and metrics for assessing adaptation effectiveness, social appropriateness, transparency, and trust across users and contexts. By bringing together researchers in multimodal AI, robotics, and HRI, it aims to identify principles and evaluation practices for robots that adapt meaningfully to people while supporting human agency.
References:
- [1] Rossi, S., Ferland, F., & Tapus, A. (2017). User profiling and behavioral adaptation for HRI: A survey. Pattern Recognition Letters, 99, 3–12.
- [2] Vinciarelli, A., Pantic, M., & Bourlard, H. (2009). Social signal processing: Survey of an emerging domain. Image and Vision Computing, 27(12), 1743–1759.
- [3] Hancock, P. A., Billings, D. R., Schaefer, K. E., Chen, J. Y. C., de Visser, E. J., & Parasuraman, R. (2011). A meta-analysis of factors affecting trust in human-robot interaction. Human Factors, 53(5), 517–527.
TOPICS
We are seeking papers on two complementary aspects of multimodal AI for human–robot interaction: understanding human social signals, states, and intentions, and translating this understanding into context-aware, socially adaptive robot behaviour. We strongly encourage submissions on the design, development, and evaluation of robots that use social reasoning, contextual awareness, and human feedback to personalise interactions, with particular attention to trust, transparency, and responsible adaptation.
Accepted topics include, but are not limited to, the following:
- Multimodal social signal understanding
- Human state and intention modelling
- Multimodal fusion and grounding
- Socially adaptive robot behaviour
- Personalisation and continual learning
- Foundation models for social robotics
- Social reasoning and cognitive architectures
- Trust, transparency, and responsible adaptation in HRI
- Human feedback and interactive learning in HRI
- Context and situation awareness for robots
- Machine Learning for robotic personalisation
- Evaluation metrics for adaptive robotic behaviour
PROGRAM CO-CHAIRS
- Alessandra Rossi, University of Naples Federico II – alessandra.rossi@unina.it
- Luigi D’Arco, University of Naples Federico II – luigi.darco@unina.it
- Luca Raggioli, University of Naples Federico II – luca.raggioli@unina.it
IMPORTANT DATES
- Extended Abstract Submission Deadline: January 31, 2027
- Notification of Acceptance: February 28, 2027
- Camera-ready Deadline: April 30, 2027
- Workshop Dates: June 9–11, 2027
PUBLICATION AND PAGE LIMITATIONS
- Contributions will be published on the SPRINGER “Smart Innovation Systems and Technologies” (see Springer Guidelines for formatting guidelines and templates). The minimum paper length is 15 pages.
- Manuscripts should be submitted through [Submission LINK TBA] by selecting “From Social Signals to Socially Adaptive Robots: Multimodal AI for Human–Robot Interaction” from the dropdown menu.
REGISTRATION FEE
Registration must be done through bank transfer and receipts must be exhibited at the conference.
Details can be found at
https://siren.science/registration.html