2/3/25

Anshuman Suri (Northeastern University): White-box vs Black-box: Privacy Auditing for Machine Learning (abstract)

2/10/25

Sahar Abdelnabi (Microsoft): Evaluating and Securing LLM-Agentic Networks (abstract)

2/20/25

Javier Rando (ETH Zurich): Gradient-based Jailbreak Images for Multimodal Fusion Models (abstract)

2/24/25

Norman Mu (xAI): A Closer Look at System Prompt Robustness (abstract)

3/3/25

Harsh Chaudhari (Northeastern University): Propagation of Adversarial Bias to Distilled Language Models (abstract)

3/10/25

Andy Zou (Carnegie Mellon Univerosity): Red Teaming AI Agents in-the-wild: Revealing Deployment Vulnerabilities (abstract)

3/24/25

Xiangyu Qi (OpenAI): Safety Alignment Should Be Made More Than Just A Few Tokens Deep (abstract)

3/31/25

Om Thakkar (OpenAI): Privacy Leakage in Speech Models: Attacks and Mitigations (abstract)

4/7/25

Ryan McKenna (Google): Private Analytics and Learning at Google (abstract)

4/10/25

Milad Nasr (Google): Topic TBD

4/18/25

Edoardo Debenedetti (ETH Zurich): Defeating Prompt Injections by Design

4/28/25

Jonas Geiping (ELLIS Institute): Topic TBD

5/5/25

Speaker TBD (TBD): Topic TBD