Sangdon Park

Online Conformal Prediction with Adversarial Semi-bandit Feedback via Regret Minimization

Ensuring Functional Correctness of Large Code Models with Selective Generation

Online Selective Generation with Adversarial Bandit Feedback

LLM Watermark Evasion via Bias Inversion

AI Kill Switch for Malicious Web-based LLM Agents

Generalization Bound for GNNs on Transductive Node Classification: A View from Optimal Transport

Holistic Unlearning Benchmark: A Multi-Faceted Evaluation for Text-to-Image Diffusion Model Unlearning

RETRIEVAL-AUGMENTED GENERATION WITH ESTIMATION OF SOURCE RELIABILITY

Deferred Backdoor Functionality Attacks on Deep Learning Models