Understanding Emergent Misalignment via Feature Superposition Geometry.
Gouki Minegishi, Hiroki Furuta, Takeshi Kojima, Yusuke Iwasawa, Yutaka Matsuo
Browse the full ACL paper archive.
Gouki Minegishi, Hiroki Furuta, Takeshi Kojima, Yusuke Iwasawa, Yutaka Matsuo
Browse the full ACL paper archive.