Renjie Pi
13 papers · 775 citations
Most-cited papers
- Mitigating The Alignment Tax Of RLHF2023 · 166 citations
- Mllm-protector: Ensuring Mllm's Safety Without Hurting Performance2024 · 118 citations
- LISA: Layerwise Importance Sampling For Memory-efficient Large Language Model Fine-tuning2024 · 106 citations
- Gradsafe: Detecting Jailbreak Prompts For Llms Via Safety-critical Gradient Analysis2024 · 95 citations
- Strengthening Multimodal Large Language Model With Bootstrapped Preference Optimization2024 · 86 citations
Topics