In Conjunction with IEEE BigData 2025
Dec 8th-11th, 2025, Macau, China
Knowledge-Augmented Multimodal Information Processing (KAMIP) aims to convene a diverse group of researchers and practitioners within the IEEE BigData community. By integrating and processing information from various modalities like text, image, audio, video, and EEG, multimodal learning has achieved remarkable success, particularly in the era of large language models (LLMs). However, existing systems often struggle with deep semantic understanding, robust reasoning, and explainability, especially in complex or data-scarce scenarios. The explicit integration of external knowledge, ranging from structured knowledge bases and ontologies to commonsense reasoning and domain-specific expertise, offers a promising avenue to overcome these limitations. This workshop will explore the critical intersection of knowledge representation, reasoning, and multimodal machine learning. We anticipate that leveraging explicit and implicit knowledge can significantly enhance the capabilities of multimodal systems, leading to more intelligent, interpretable, and trustworthy AI across a wide array of applications, from advanced content analysis and human-computer interaction to decision support systems. Through this workshop, we aim to foster discussion and showcase cutting-edge research on novel theories, methodologies, and applications that effectively infuse knowledge into the fabric of multimodal information processing.
Knowledge Discovery from Multimodal Data
Knowledge Representation from Multimodal Data
Knowledge Extraction from Multimodal Sources
Interactive Multimodal Knowledge Discovery
Knowledge-Augmented Multimodal Learning
Models and Techniques for Knowledge-Augmented Multimodal Learning
Knowledge-Augmented Multimodal Large Language Models
Zero-shot/Few-shot Multimodal Learning with Augmented Knowledge
Neuro-Symbolic Approaches for Multimodal AI
Knowledge-Augmented Multimodal Reasoning
Knowledge-Augmented Multimodal Generation
Construction and Applications of Multimodal Knowledge Graph Embeddings
Applications of Knowledge-Augmented Multimodal Processing
Datasets, Benchmarks, and Evaluation Metrics
Explainable and Interpretable Knowledge-Augmented Multimodal Systems
Ethics and Bias in Multimodal AI
Fairness, Transparency, and Societal Impact of Multimodal Systems
Due Date for Full Papers Submission: October 19th, 2025
Notification of Paper Acceptance: November 9th, 2025
Camera-ready Paper Deadline: November 23rd, 2025