Poster · Sydney
Do You CARE to Generalize? Extracting Robust Concept Directions from LLMs
- When
- Tue 8 Dec 2026, 17:00–20:00 AEDT
- Where
- Hall 1-4, ICC Sydney
Exhibition Centre, Level 1 - Session
- Sydney Poster Session 2
- Authors
- Sweta Karlekar, Claudia Shi, Aahlad Manas Puli, Carolina Zheng, Maggie Makar, John Bowlan, Michal Kucer, David Blei
- Organisations
- Columbia University, New York University, University of Michigan, Los Alamos National Laboratory
- QRS classification
- Language models