MAPLE2025 @ ICMLA 2025

Introduction

This special session aims to highlight cutting-edge research in the development and practical deployment of multimodal machine learning algorithms that integrate and process heterogeneous data types—such as text, image, audio, video, and sensor signals—to address complex, real-world challenges. By leveraging the complementary strengths of multiple modalities, these systems enable more robust, context-aware, and intelligent solutions across a wide range of domains, including healthcare, cybersecurity, robotics, smart environments, transportation, surveillance and so on. The session invites contributions where algorithmic innovations in multi-modal AI systems are developed in the context of real-world applications, leading to tangible improvements in performance, scalability, and interpretability. We particularly welcome work demonstrating multi-modal AI in domains such as healthcare, cybersecurity, robotics, smart homes, transportation, and surveillance. We also encourage submissions focusing on educational tools, mobile and web-based AI systems, multi-modal chatbot development, and cross-modal retrieval tasks.

Scopes

This workshop will cover but not limit to the following topics:

Vision-language and multi-modal foundation models
Generative models for multi-modal synthesis
Multi-modal representation alignment and fusion techniques
Transfer learning and fine-tuning strategies in multi-modal deep learning
Cross-modal retrieval and matching (e.g., image-to-text, audio-to-video)
Domain adaptation and self-supervised learning for multi-modal data
Explainable, interpretable, and trustworthy multi-modal ML systems
Applications in cybersecurity, medical imaging, transportation, robotics, and smart environments
Mobile, web, and edge deployment of multi-modal systems
Real-time architectures and lightweight multi-modal models for deployment
Benchmark datasets, framework, and reproducibility in multi-modal ML

Submission Guidelines and Instructions

Papers submitted for reviewing should conform to IEEE specifications. Manuscript templates can be downloaded from IEEE website
The maximum length of papers is 8 (eight) pages.
All the papers will go through double-blind peer review process.
Authors’ names and affiliations should not appear in the submitted papers.
Authors’ prior work should be cited in the third person.
Authors should take care and avoid revealing their identities and/or institutions in the paper’s text, figures, links, etc.
Any papers that do not adhere to the double-blind peer review policy will be rejected without peer review.

Submission Platform: Submit via the CMT Submission Site*.
Select the track: Special Session: Multi-modal Machine Learning in Practice: Algorithms and Applications (MAPLE2025)
*CMT ACKNOWLEDGMENT: "The Microsoft CMT service was used for managing the peer-reviewing process for this conference. This service was provided for free by Microsoft and they bore all expenses, including costs for Azure cloud services as well as for software development and support."