2026

论文总结-SafeGRPO:Self-Rewarded Multimodal Safety Alignment via Rule-Governed Policy Optimization 03-30 论文总结-Safe RLHF-V:Safe Reinforcement Learning from Multi-modal Human Feedback 03-29 论文总结-Adversary-Aware DPO:Enhancing Safety Alignment in Vision Language Models via Adversarial Training 03-24