LoX: Low-Rank Extrapolation Robustifies LLM Safety Against Fine-tuning

LoX: Low-Rank Extrapolation Robustifies LLM Safety Against Fine-tuning

18 June 2025

Gabrel J. Perin

Runjin Chen

Xuxi Chen

Nina S. T. Hirata

Zhangyang Wang

Junyuan Hong

ArXiv (abs)PDF HTML

Papers citing "LoX: Low-Rank Extrapolation Robustifies LLM Safety Against Fine-tuning"

Title
No papers