Symmetry and Asymmetry in Next-Generation Artificial Intelligence: Unimodal, Multimodal, and Generative Models
This special issue belongs to the section "A: Computer Science".
Special Issue Information
Dear Colleagues,
Artificial intelligence is undergoing a rapid transformation driven by the development of large-scale learning models, generative artificial intelligence, foundation models, intelligent agents, and increasingly sophisticated unimodal and multimodal systems. These approaches are being applied across a wide range of scientific, technological, industrial, environmental, and societal domains. However, their growing complexity also introduces important challenges related to consistency, invariance, representation, data imbalance, heterogeneous information, model robustness, and the interaction between multiple sources, modalities, and intelligent components.
Symmetry and asymmetry provide a valuable framework for studying many of these challenges. Symmetry may appear in artificial intelligence through invariant or equivariant representations, consistent model behavior under transformations, balanced information processing, structural regularities, or equivalent representations of semantically similar inputs. Conversely, asymmetry may arise from heterogeneous data distributions, imbalanced classes, differences between modalities, unequal information availability, asymmetric interactions among agents, domain shifts, or variations in model performance across populations, tasks, languages, and environments.
Recent advances in unimodal, multimodal, and generative artificial intelligence provide new opportunities to investigate how these properties can be modeled, preserved, exploited, or evaluated. Unimodal systems continue to provide important methodological advances in language, vision, audio, time-series analysis, and structured data processing. Multimodal approaches seek to integrate heterogeneous sources of information, while generative models—including large language models, vision–language models, diffusion models, and foundation models—are enabling increasingly general and adaptive intelligent systems. Intelligent agents based on these models further introduce new interactions between reasoning, memory, external tools, data sources, and autonomous decision-making.
This Special Issue aims to bring together theoretical, methodological, and application-oriented research addressing symmetry and asymmetry in next-generation artificial intelligence. We welcome original research and review articles exploring how symmetry-related principles can improve the design, analysis, evaluation, robustness, interpretability, and application of modern artificial intelligence systems.
The topics of interest include the following:
- Symmetry and asymmetry in artificial intelligence and machine learning;
- Unimodal learning in text, image, audio, video, sensor, time-series, and structured data;
- Multimodal learning and multimodal information fusion;
- Generative artificial intelligence and generative models;
- Large language models and vision–language models;
- Foundation models and domain-specific foundation models;
- LLM-based and multimodal intelligent agents;
- Multi-agent artificial intelligence systems;
- Retrieval-augmented generation and knowledge-enhanced AI;
- Equivariant and invariant neural networks;
- Symmetry-aware and symmetry-preserving deep learning;
- Geometric deep learning;
- Representation learning and transformation invariance;
- Semantic consistency and robustness under equivalent inputs;
- Data imbalance and asymmetric data distributions;
- Cross-modal symmetry and asymmetry;
- Domain adaptation, transfer learning, and domain shifting;
- Explainable and trustworthy artificial intelligence;
- Robustness, uncertainty, and reliability of generative models;
- Bias, fairness, and asymmetric model behavior;
- Synthetic data generation and augmentation;
- AI agents interacting with external tools and knowledge sources;
- Scientific and engineering applications of next-generation AI;
- Artificial intelligence for healthcare, agriculture, environment, energy, manufacturing, smart cities, and autonomous systems;
- Novel evaluation frameworks for unimodal, multimodal, and generative AI.
By integrating symmetry- and asymmetry-related perspectives with emerging artificial intelligence paradigms, this Special Issue seeks to provide a multidisciplinary forum for advances that contribute to more robust, interpretable, adaptive, and generalizable intelligent systems.
Dr. Alfonso Ramirez-Pedraza
Dr. José Joel González-Barbosa
Guest Editors
Manuscript Submission Information
Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the Special Issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.
Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-anonymized peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Symmetry is an international peer-reviewed open access monthly journal published by MDPI.
Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.
Keywords
- symmetry
- asymmetry
- artificial intelligence
- generative AI
- large language models
- multimodal learning
- unimodal learning
- foundation models
- intelligent agents
- equivariant learning
Benefits of Publishing in a Special Issue
- Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
- Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
- Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
- External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
- Reprint: MDPI Books provides the opportunity to republish successful Special Issues in book format, both online and in print.

