AI RESEARCH
CrispEdit: Low-Curvature Projections for Scalable Non-Destructive LLM Editing
arXiv CS.LG
•
ArXi:2602.15823v2 Announce Type: replace A central challenge in large language model (LLM) editing is capability preservation: methods that successfully change targeted behavior can quietly game the editing proxy and corrupt general capabilities, producing degenerate behaviors reminiscent of proxy/reward hacking. We present CrispEdit, a scalable and principled second-order editing algorithm that treats capability preservation as an explicit constraint, unifying and generalizing several existing editing approaches.