git-disl/Booster

This is the official code for the paper "Booster: Tackling Harmful Fine-tuning for Large Language Models via Attenuating Harmful Perturbation" (ICLR2025 Oral).

语言：Shell · GitHub 星标：43 · Fork：2

话题：alignment · attack · emergent · fine-tuning · finetuning · harmful · llms · malicious

在 GitHub 上查看原仓库