Trustworthy Decision Intelligence: Data-Centric AI, Foundation Models, and Real-World Impact

Dear Colleagues,

This Special Issue focuses on the fast-moving shift from “model-centric” research toward decision-oriented, trustworthy AI systems that can be deployed in real services. Recent progress in foundation models, retrieval-augmented generation and privacy-aware collaborative learning is opening new ways to connect heterogeneous data with robust inference and actionable recommendations. At the same time, real deployments expose hard constraints—limited labels, non-IID data, communication overhead and the need for transparency—that demand methods going beyond incremental accuracy improvements.

We invite submissions across a broad range of contemporary topics, including (but not limited to) LLM-enabled decision support (grounding, retrieval, tool use, multimodal reasoning), efficient and resilient learning (federated and semi-supervised learning, personalization under heterogeneity, compression and communication efficiency, edge–cloud co-optimization) and trustworthy evaluation (privacy/security, robustness, calibration and uncertainty, interpretability, reproducibility). We particularly welcome studies that link technical innovations to clear operational benefits, such as reduced cost, improved reliability, lower human workload, or better decision quality under realistic constraints.

The purpose of this Special Issue is to bring together cutting-edge research that explicitly connects data → learning → decision → impact. A key message is that real-world success is often defined by the effectiveness of downstream actions, not by a single headline metric. For instance, in education analytics, many universities build dropout predictors, yet the practical objective is maximizing rescue outcomes with limited counseling capacity. Decision-aware strategies—such as combining a high-recall model to secure sufficient at-risk coverage with a high-precision model to shrink the intervention list—illustrate how AI can increase rescue rates while controlling additional effort. This “intervention-first” framing generalizes naturally to domains such as healthcare triage, energy scheduling and security operations.

By bringing together conference-linked and open submissions, this Special Issue will complement existing literature by emphasizing SOTA methods with deployment-grounded evaluation and by presenting a unified perspective on trustworthy decision intelligence. It also resonates with the broader conversations in the PlatCon community, while extending them toward the newest trends in foundation models, data-centric learning and actionable, reliable AI for real services.

Dr. Jihoon Moon
Dr. Jehyeok Rew
Dr. Hyeon-Woo Kim
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. Electronics 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 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.

• trustworthy AI
• decision intelligence
• data-centric learning
• foundation models
• large language models
• retrieval-augmented generation
• semi-supervised learning
• federated learning
• non-IID robustness
• privacy-preserving analytics
• uncertainty and calibration
• explainable AI
• cost-sensitive optimization
• human-in-the-loop
• educational analytics
• dropout intervention
• edge–cloud efficiency