The Geometry of Refusal in Large Language Models: Concept Cones and Representational Independence.
Tom Wollschlger, Jannes Elstner, Simon Geisler, Vincent Cohen-Addad, Stephan Gnnemann, Johannes Gasteiger
Browse the full ICML paper archive.
Tom Wollschlger, Jannes Elstner, Simon Geisler, Vincent Cohen-Addad, Stephan Gnnemann, Johannes Gasteiger
Browse the full ICML paper archive.