Workshop Goals

As foundation models become increasingly integrated into critical applications, ensuring their trustworthiness is paramount. The International Symposium on Trustworthy Foundation Models brings together researchers, industry leaders, and policymakers to discuss key challenges and advancements in building safe, fair, and robust AI systems. This symposium will explore topics such as model safety, privacy, fairness, interpretability, causality, and robustness, with a particular emphasis on real-world deployment and governance. Through two days of invited talks, networking discussions, and contributed research, we aim to foster collaboration and drive innovation in the responsible development of foundation models.

Areas of Focus

Specifically, the objectives of the symposium are

• Exchanging ideas, e.g., we will invite keynote speeches from world-leading researchers and contributed talks from active researchers;

• Building networks and promoting potential collaborations, e.g., promote internal collaborations within MBZUAI and internal collaborations with rising-star and world-leading researchers;

• Inspiring the juniors to do better research, e.g., mentor research students and postdocs at MBZUAI to do high-qulity research;

• Improving MBUZAI’s profile and attracting rising stars to join us;

Schedule

We will host two days of invited talks (May 26-27) interspersed with panel discussions on challenges in learning with large models. We will also organize a research night for the students to discuss with speakers.

This will be followed by transfers to the airport for departing flights.

Organizers

Organizing committee:
General chair: Kun Zhang, MBZUAI
Program chair: Tongliang Liu, MBZUAI/USYD; Bo Li, UIUC
Invited faculty talk session chair: Nils Lukas, MBZUAI; Bo Han, HKBU/RIKEN
Rising-star presentation session chair: Salem Lahlou, MBZUAI
PhD mentoring session chair: Mingming Gong, MBZUAI/UoM
Local arrangement chair: Runqi Lin, MBZUAI/USYD