Yuhang Lai

code generation (2) large language model (2) benchmark evaluation (1) adversarial robustness (1) reward modeling (1) preference alignment (1) language model alignment (1) model safety (1) knowledge base (1) reinforcement learning from human feedback (1) preference modeling (1) human feedback (1) ensemble method (1) reward model (1) retrieval-augmented generation (1) large vision-language model (1) human preference (1) jailbreak defense (1) token-level prediction (1) functional correctness (1)