Huijie Guo

bi-level optimization (2) self-supervised learning (2) feature learning (1) reward hacking (1) sparse autoencoder (1) causal analysis (1) backdoor adjustment (1) process reward model (1) task conflict (1) semantic invariance (1) meta comprehensive regularization (1) maximum entropy coding (1) causal inference (1) representation transferability (1) representation learning (1)