Skip to content
NeurIPS 2026 planner

Poster · Sydney

AnchorRep: Defending LLMs Against Cross-Model Adversarial Transfer via Representation Repulsion

When
Wed 9 Dec 2026, 10:00–13:00 AEDT
Where
Hall 1-4, ICC Sydney
Exhibition Centre, Level 1
Session
Sydney Poster Session 3
Authors
Gal Wertheizer, Rom Himelstein, Tomer Peretz, Avi Mendelson
Organisations
Technion
QRS classification
Language models, Safety, fairness and privacy

Open in scheduleOfficial page