Sihong Xie

retrieval-augmented generation (2) faithfulness hallucination (2) multi-armed bandit (2) adversarial training (2) adversarial robustness (1) maximum entropy (1) metric learning (1) uncertainty quantification (1) certified robustness (1) data augmentation (1) rumor detection (1) question answering (1) knowledge graph question answering (1) label accuracy (1) inverse reinforcement learning (1) feature attribution (1) exploration-exploitation tradeoff (1) closed-form solution (1) face verification (1) bayesian inference (1)