WORKSHOP ON RELIABLE AND PREDICTIVE GENERATIVE AI (IJCNN 2026) Link: https://cmt3.research.microsoft.com/DDMGenAI2026

--- CALL FOR PAPERS ---
Generative AI (GenAI) has seen tremendous advances driven by deep learning, evolving from early energy-based models to expressive frameworks such as score-based, diffusion, and flow-based models. These paradigms have demonstrated remarkable capabilities in learning complex data distributions and capturing underlying structures in high-dimensional spaces. However, the potential of modern generative models extends far beyond conventional data synthesis. Their probabilistic nature offers powerful tools for forecasting, inverse problem-solving, control, and scientific discovery. By providing Bayesian interpretations and probabilistic predictions, these models enable uncertainty-aware and data-driven decision-making, crucial for safety-critical and real-world domains.
This workshop aims to explore advances in reliable generative modeling. We focus on methods for uncertainty quantification, robustness under distributional shift, and the calibration of probabilistic outputs. We particularly encourage interdisciplinary contributions that bridge deep generative modeling, neural network theory, and probabilistic inference.

The workshop emphasizes two core pillars:
- Beyond Synthesis: Showcasing how GenAI applies to decision-making, complex systems modeling, scientific computing, robotics, signal processing, and health.

- Predictive Capability: Exploring uncertainty quantification, probabilistic reasoning, and the Bayesian foundations that support informed, trustworthy decision-making.

--- TOPICS OF INTEREST ---
We invite submissions that reinforce the reliability of Generative AI in real-world applications. Topics include, but are not limited to:

Foundations: Score-based, flow-based, and diffusion generative methodologies; Probabilistic and Bayesian generative modeling; Neural network theory for GenAI.

Reliability & Trust: Generative AI for trustworthy, explainable, and interpretable machine learning; Robustness under distributional shift and concept drift.

Decision Making & Control: Uncertainty-aware reinforcement learning; Generative control; GenAI for data-driven decision-making.

Applications: Application of GenAI beyond standard data synthesis (e.g., scientific discovery, robotics, inverse problems); Cross-disciplinary real-world deployments.

Evaluation: Metrics, benchmarks, and reproducibility in generative AI.

--- SUBMISSION GUIDELINES ---

Submissions must follow the IJCNN 2026 formatting rules (available at: https://attend.ieee.org/wcci-2026/information-for-authors/). We invite two types of contributions:

Full Papers: Up to 8 pages.

Short Papers: Up to 4 pages.

All papers must be submitted via the Microsoft CMT portal: https://cmt3.research.microsoft.com/DDMGenAI2026