Jun Sakuma

Toward Safer Diffusion Language Models: Discovery and Mitigation of Priming Vulnerability

Learning Semi-Structured Sparsity for LLMs via Shared and Context-Aware Hypernetwork

Robust Deep Reinforcement Learning against Adversarial Behavior Manipulation

Rewriting Pre-Training Data Boosts LLM Performance in Math and Code

Model-agnostic Adversarial Attack and Defense for Vision-Language-Action Models

Robust Deep Reinforcement Learning against ADVERSARIAL BEHAVIOR MANIPULATION