ITGenNexus: Bridging Information Theory and Generative AI

Overview

ITGenNexus is a one-day ISIT 2026 workshop focused on the emerging interface between Information Theory and Generative AI. The workshop aims to foster collaboration between these two communities by promoting dialogue on how information-theoretic principles can deepen our understanding of modern generative models, and how the rapid advances in generative AI (GenAI) can, in turn, inspire new directions and challenges in information theory.

Recent progress has demonstrated the power of information-theoretic tools in analyzing and enhancing the robustness, efficiency, and trustworthiness of machine learning and GenAI systems. These efforts span a wide range of topics, including watermarking and attribution of large language models, generalization and reasoning mechanisms, robustness under distribution shifts, and multi-modal representation learning that integrates text, vision, and audio. At the same time, the unprecedented problem-solving capabilities of GenAI systems have exposed long-standing “black boxes” in learning and inference, motivating the development of new theoretical frameworks and analytical tools.

The workshop will bring together researchers from information theory, machine learning, and AI through invited talks by leading experts from academia and industry, complemented by contributed presentations on both theoretical foundations and practical applications. By creating a shared forum for these communities, ITGenNexus aims to equip AI researchers with principled information-theoretic perspectives, while enabling traditional ISIT participants to engage deeply with modern generative models and their open challenges. Ultimately, the workshop seeks to catalyze new collaborations and advance the theoretical and practical understanding of generative AI.

Key information

• Paper Format: Same as ISIT main conference (non-archival)
• Workshop Format: Invited Talks + Contributed Presentations + Poster & Panel Session

Organizing committee

YL

Yingbin Liang

The Ohio State University

JJ

Jiantao Jiao

University of California, Berkeley

YB

Yuheng Bu

University of California, Santa Barbara

HH

Haiyun He

The Hong Kong University of Science and Technology (Guangzhou)

ZW

Ziqiao Wang

Tongji University

PW

Peng Wang

University of Macau