Shujian Zhang

DLM-One: Diffusion Language Models for One-Step Sequence Generation

Eliciting Behaviors in Multi-Turn Conversations

Fantastic Reasoning Behaviors and Where to Find Them: Unsupervised Discovery of the Reasoning Process

Score Forgetting Distillation: A Swift, Data-Free Method for Machine Unlearning in Diffusion Models

Instructional Segment Embedding: Improving LLM Safety with Instruction Hierarchy

Statistical Advantages of Perturbing Cosine Router in Mixture of Experts

SFTMix: Elevating Language Model Instruction Tuning with Mixup Recipe