Poster · Sydney
Taming CoT Obfuscation in VLMs: From Mechanistic Evidence to Activation-Level Enforcement
- When
- Wed 9 Dec 2026, 17:00–20:00 AEDT
- Where
- Hall 1-4, ICC Sydney
Exhibition Centre, Level 1 - Session
- Sydney Poster Session 4
- Authors
- Xutao Mao, Jianing Zhu, Jinman Zhao, Tongliang Liu, Xiaowen Chu, Cong Wang, Bo Han
- Organisations
- City University of Hong Kong, University of Texas at Austin, University of Toronto, Mohamed bin Zayed University of Artificial Intelligence, Hong Kong University of Science and Technology (Guangzhou), Hong Kong Baptist University, RIKEN
- QRS classification
- Reinforcement learning, Language models