Viet Anh Nguyen

Adaptive Rollout Allocation for Online Reinforcement Learning with Verifiable Rewards

Exploring Diverse Generation Paths via Inference-time Stiefel Activation Steering

Scaling Language Model Reliability via Determinantal Point Process Prompt Sampling

Reasoning Planning for Language Models

Risk-Aware Distributional Intervention Policies for Language Models

Forget but Recall: Incremental Latent Rectification in Continual Learning