Search papers, labs, and topics across Lattice.
University of Pavia, Cochin University of Science and Technology
2
0
3
Current safety mechanisms for LLMs are critically flawed, as a genetic algorithm can effectively evolve adversarial prompts to bypass them entirely.
Forget retraining: NeWTral instantly restores safety to your LLM after adding a risky LoRA, slashing attack success rates from 70% to 13% without sacrificing expertise.
