Thanh Nguyen-Tang

Policy Regret Minimization in Partially Observable Markov Games

How Transformers Learn In-Context Recall Tasks? Optimality, Training Dynamics and Generalization

Constraint-Aware Reward Relabeling for Offline Safe Reinforcement Learning

S-Chain: Structured Visual Chain-of-Thought for Medicine

Exact Unlearning in Reinforcement Learning

Online Optimization for Offline Safe Reinforcement Learning

Policy-Regret Minimization in Markov Games with Function Approximation

Wicked Oddities: Selectively Poisoning for Effective Clean-Label Backdoor Attacks

Neural ODE Transformers: Analyzing Internal Dynamics and Adaptive Fine-tuning

Update larger, train faster: Stable Test-time adaptation utilizing noisy-pseudo labels