Lei Hsiung

representation learning (1) adversarial robustness (1) domain adaptation (1) backdoor defense (1) adversarial training (1) energy efficiency (1) safety alignment (1) spectral analysis (1) model inference (1) probability calibration (1) adversarial attack (1) adversarial defense (1) adversarial example (1) jailbreak attack (1) robustness evaluation (1) neural network inference (1) confidence estimation (1) model accuracy (1) neural network calibration (1) centered kernel alignment (1)