Latent Adversarial Regularization for Offline Preference Optimization
Paper • 2601.22083 • Published • 14
AI Safety and Alignment
-
-
RAMP: Boosting Adversarial Robustness Against Multiple l_p Perturbations for Universal Robustness
Paper • 2402.06827 • Published -
Misaligning Reasoning with Answers -- A Framework for Assessing LLM CoT Robustness
Paper • 2505.17406 • Published -
Towards Universal Certified Robustness with Multi-Norm Training
Paper • 2410.03000 • Published -
EnyiJiang/Gemma-2-2B-it-GANPO
0.7B • Updated • 5 -
EnyiJiang/Llama-3-8B-Instruct-GANPO
2B • Updated • 8