Towards Reliable, Trustworthy and Privacy-Preserving Mathematical Modeling and Data Readiness for Foundation Models with Industrial Applications

Dear Colleagues,

Recently, Foundation Models (FMs), such as large language models, vision Foundation Models, and time series Foundation Models, have shown unprecedented development and remarkable performance in various tasks and application fields, thus demonstrating their potential to drive new technological revolutions and be used in advanced applications in the industrial domain. However, due to domain-specific challenges and data confidentiality issues in the industrial sector, there exist significant limitations in harnessing the capabilities of Foundation Models. Such challenges have required the study of modeling paradigms and heterogeneous architectures for Foundation Models (FMs) with industrial applications, demanding novel exploration and the guarantee of resilient and dependable cohesion between theoretical algorithms and real-world engineering systems. For this reason, we wish to draw special attention to the issues of mathematical modeling and data readiness for Foundation Models with industrial applications, aiming to develop original solutions to the aforementioned challenges facing them.

In this Special Issue, we aim to include a wide spectrum of research papers covering the relevant reliability, trustworthiness, and privacy issues in the mathematical modeling and data acquisition process of FMs with industrial applications, such as challenges related to their data readiness, data quality evaluation, data privacy, critical data protection, data integration and aggregation, and model resilience and generalization. We welcome high-quality research from both theory and application perspectives and the discussion of methodologies ranging from the use of FM training technologies to efficient mathematical modeling theories to promote academic exchange between a wide array of scholars. As such, we invite you to submit an article to this peer-reviewed Special Issue, “Towards Reliable, Trustworthy and Privacy-Preserving Mathematical Modeling and Data Readiness for Foundation Models with Industrial Applications”. It will focus on (but not be limited to) the following topics:

• Data acquisition, processing, preparation, and anonymization for foundation models;
• Data integration and aggregation for foundation models;
• Mathematical modeling to achieve reliability, trustworthiness, and privacy in foundation models;
• Mathematical solutions for data synthesis, data augmentation, data selection, and data mixing;
• Mathematical solutions for data quality evaluation, data valuation, data pricing, and the data market;
• Privacy preservation and confidentiality protection for foundation models;
• Poisoning, backdoor attacks, adversarial attacks, and jailbreaking data detection, filtering, and cleaning;
• Advanced threat models: cybercrime, cyber-espionage, and security vulnerabilities;
• Designing reliable, trustworthy, and private foundation models;
• Data exploration and wrangling for the IoT;
• Multimodal data readiness for multimodal foundation models.

Prof. Dr. Zibin Zheng
Dr. Dan Li
Dr. Jian Lou
Dr. See-Kiong Ng
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 100 words) can be sent to the Editorial Office for announcement on this website.

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-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Mathematics is an international peer-reviewed open access semimonthly 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 2600 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.

• Data readiness
• Foundation model
• Reliable and trustworthy AI
• Data synthesis, data augmentation, and data selection
• Data privacy
• Data quality, data valuation, and data pricing
• Adversarial machine learning
• Efficient model architecture