Event schedules / ICONIP 2026

SS-02: Multimodal Learning and Digital Twins for Advancing Affective Human-Machine Interaction

Monday, November 23, 2026, – (Australia/Sydney)
Clarendon Auditorium · Special Session

Multimodal Learning and Digital Twins for Advancing Affective Human-Machine Interaction

Abstract: This session explores the integration of multimodal learning and digital twins to enhance affective computing, enabling intelligent systems to better recognize, interpret, and adaptively express emotional cues. It aims to bridge the gap between theoretical advancements and practical implementations in domains such as healthcare, education, and robotics. Key topics include facial and speech emotion recognition, generative emotion modeling, and real-time affective feedback systems. Notably, selected papers from this session will be invited to submit extended versions to a special collection in Springer's Cognitive Computation journal.

Note: Extended versions of the selected papers in this session will be invited to publish in a special issue of the Cognitive Computation journal by Springer.

Paper 200 - MemEvo:Towards Human- AI Co-Adaptation through Evolving Memory in Mobile Task Automation (V)
Authors: Zhao Peiling

Paper 988 - U-VFL: A Unified Framework for Secure Vertical Federated Learning with Adaptive Privacy and Verifiable Computation
Authors: Md Imran Ahmed, Esika Arifin Rumky, Taha Osman, Mufti Mahmud, David J. Brown and Muhammad Arifur Rahman

Paper 993 - Interpretable Federated Learning for Alzheimer’s Disease: Evaluating Explainable AI Across Distributed Clinical Data
Authors: Danruo Zhang, Muhammad Arifur Rahman, Taha Osman, Nicholas Shopland, Mufti Mahmud and David J. Brown

Paper 208 - Tuning the Eyes and Brain: How Does Structural Fine-Tuning Adapt Semantic VLMs for Robotics?
Authors: Dongxiao Hu, Yichen Luo, Chencheng Wang, Anda Liu, Jia Wang, Dapeng Lan, Yu Liu and Zhibo Pang