Call for Papers
Multi-scale Artificial Intelligence (MAI) focuses on the development of AI methods, models, systems, and applications that explicitly learn, represent, reason about, or integrate information across multiple scales. These scales may be spatial, temporal, semantic, computational, organizational, or hierarchical, depending on the research problem and application context.
By integrating information and intelligence across different levels of granularity, MAI aims to capture cross-scale relationships, interactions, and dependencies that may not be adequately represented at a single scale. This perspective is relevant to a broad range of AI research, including multi-scale representation learning, hierarchical and multi-resolution modeling, cross-scale learning and reasoning, multi-scale perception, multi-scale decision-making, and intelligent systems operating across different levels of abstraction.
MAI is particularly relevant to complex real-world problems in which information, processes, and decisions span multiple scales. Applications may include computer vision, natural language processing, robotics and autonomous systems, scientific discovery, healthcare, engineering, intelligent networks, cyber-physical systems, and other complex environments.
MAI 2027 welcomes original research that advances the theory, methodology, systems, and applications of Multi-scale Artificial Intelligence. We particularly encourage submissions in which multi-scale or cross-scale aspects constitute a substantive component of the research problem, methodology, model, system, or application.
| Track 1: Foundations of Multi-scale Artificial Intelligence |
Multi-scale Machine Learning, Multi-scale Representation Learning, Multi-scale Deep Learning, Multi-scale Neural Networks, Multi-scale Optimization and Learning, Multi-scale Probabilistic Modeling, Multi-scale Reinforcement Learning, Multi-scale Federated and Distributed Learning, Multi-scale Graph Learning, Multi-scale Knowledge Representation, Multi-scale Reasoning and Decision Making, Multi-scale Explainable and Trustworthy AI, Multi-scale Continual and Adaptive Learningng |
| Track 2: Multi-scale Perception, Intelligence and Systems |
Multi-scale Computer Vision, Multi-scale Object Detection and Recognition, Multi-scale Multimodal Learning, Multi-scale Natural Language Processing, Multi-scale Large Language Models, Multi-scale Generative AI, Multi-scale Intelligent Agents, Multi-scale Robotics and Autonomous Systems, Multi-scale Human-AI Interaction, Multi-scale Cyber-Physical Systems, Multi-scale Intelligent Networks, Multi-scale Edge and Distributed Intelligence |
| Track 3: Multi-scale AI for Complex Systems and Applications |
Multi-scale AI for Healthcare and Medicine, Multi-scale AI for Scientific Discovery, Multi-scale AI for Engineering and Manufacturing, Multi-scale AI for Smart Cities, Multi-scale AI for Transportation, Multi-scale AI for Energy Systems, Multi-scale AI for Earth and Environmental Systems, Multi-scale AI for IoT, Multi-scale AI for Autonomous Systems, Multi-scale AI for Finance and Business, Multi-scale AI for Education, Multi-scale AI for Complex Decision Support |
| Track 4: Emerging Frontiers of Multi-scale AI |
Trustworthy and Responsible Multi-scale AI, Privacy-Preserving Multi-scale AI, Human-AI Collaboration, Brain-Inspired Multi-scale AI, Neuro-AI, Cognitive AI, Sustainable Multi-scale AI, AI for Scientific Discovery, Multi-scale AI Governance and Ethics, Emerging Architectures for Multi-scale AI, Cross-scale Learning and Reasoning, Multi-scale AI for Foundation Models, Multi-scale AI Agents, Open-source Multi-scale AI Systems
|
GUIDE FOR SUBMISSIONS:
*Papers prepared in the prescribed format are to be submitted. The papers should be written in English and clearly state the title, objective, method, results, and conclusion with major keywords. All papers submitted will be checked for plagiarism.
*Author names and their affiliations should be removed from the initial PDF file for the double blind review process. After receiving, the full paper will be peer-reviewed and its acceptance will be notified. Only papers presenting original content with novel research results or successful innovative applications will be considered for publication in the conference proceedings.
*The paper should be original work of the author(s), and no portion of the paper (including, but not limited to, graphics and figures) has been previously published. The paper is not currently under consideration for publication elsewhere.
*The authors listed on the paper accurately reflect those who actually did the work and contributed to the paper in a meaningful way, and they have identified and acknowledged all sources used in the creation of their paper or manuscript, including any graphics, images, tables, and figures.
PLAGIARISM POLICY:
*The paper prior to submission should be checked for plagiarism from licensed plagiarism tools like Turnitin or CrossCheck. The similarity content should not exceed 25% (in any case either self contents or others).
*Any form of self-plagiarism or plagiarism from others' works should not be there in a paper. If any model / concept / figure / table / data / conclusive comment by any previously published work is used in your paper, you should properly cite a reference to the original work.
