The 27th International Conference on Multimodal Interaction (ICMI 2025) will be held in Canberra, Australia. ICMI is the premier international forum for advancing research at the intersection of multimodal artificial intelligence (AI) and social interaction. Multimodal AI addresses technical challenges in machine learning and computational modelling, such as representations, data fusion, system design, and large-scale data management. Social interaction research encompasses both human-human interactions and human-computer interactions to uncover insights that bridge behavioural understanding with technology. A unique aspect of ICMI is its multidisciplinary nature which values both scientific discoveries and technical innovations, with an eye towards impactful applications that benefit people and society.

Novelty will be evaluated along two dimensions: scientific novelty and technical novelty. Accepted papers at ICMI 2025 must demonstrate novelty in at least one of these dimensions:

Scientific Novelty: Papers should contribute new scientific insights into human social interactions, including human-computer interactions. For example, uncovering new behavioural markers, predictive of mental health or identifying novel behavioural patterns in children’s interactions during learning. Authors are expected to conduct a thorough literature review and clearly articulate the scientific advancements and discoveries presented in their work.
Technical Novelty: Papers should introduce innovations in their computational approach for recognising, generating or modelling multimodal data. Examples of technical novelty include: advancements in learning algorithms, prediction models, neural architecture, or data representation. Novelty may also arise from creative and impactful new applications of existing methodologies.

Commitment to ethical conduct is mandatory and submissions must adhere to ethical standards in particular when human-derived data are employed. Authors are encouraged to consult the ACM Code of Ethics and Professional Conduct (https://ethics.acm.org/).

By ensuring both innovation and ethical integrity, ICMI 2025 aims to promote research that is not only technically and scientifically groundbreaking but also responsible and impactful.

ICMI 2025 conference theme

The theme of this year’s conference revolves around “Safe and Responsible Multimodal Interactions”. The focus is on exploring how multimodal systems and multimodal interactive applications can support safe, ethical, and inclusive interactions across diverse contexts, ensuring that advancements in multimodal AI and interaction technologies align with societal values and address pressing global challenges.

This theme emphasises the importance of developing systems that prioritise users’ well-being, trust, and fairness. It calls for addressing critical issues such as bias mitigation, ensuring transparency in decision-making processes, and upholding privacy and security in handling multimodal data. Researchers are encouraged to explore innovative approaches that promote inclusivity and accessibility for underrepresented communities and vulnerable populations, fostering a more equitable digital future.

Furthermore, the theme highlights the role of multimodal AI in addressing real-world safety concerns, such as building reliable systems for mental health monitoring, assisting vulnerable users (e.g., children, elderly individuals, or those with disabilities), and supporting safety-critical applications, including healthcare, education, and autonomous systems.

ICMI’25 seeks to advance the state of multimodal interaction research while maintaining a strong focus on ethical integrity, accountability, and positive societal impact. By emphasising the importance of safe and responsible design, this year’s conference aspires to pave the way for technologies that empower individuals, enhance social good, and contribute to building a more trustworthy, inclusive, and sustainable technological ecosystem.

Topics of interest include but are not limited to:

Affective computing and interaction
Cognitive modelling and multimodal interaction
Gesture, touch and haptics
Healthcare, assistive technologies
Human communication dynamics
Human-robot/agent multimodal interaction
Human-centred AI and ethics
Interaction with smart environment
Machine learning for multimodal interaction
Mobile and wearable multimodal systems
Multimodal behaviour generation
Multimodal datasets and validation
Multimodal dialogue modeling
Multimodal fusion and representation
Multimodal interactive applications
Novel multimodal datasets
Spoken/visual behaviours in social interaction
System components and multimodal platforms
Virtual/augmented reality and multimodal interaction