Shitong Shao

OD$^3$: Optimization-free Dataset Distillation for Object Detection

Late-to-Early Training: LET LLMs Learn Earlier, So Faster and Better

Guidance Matters: Rethinking the Evaluation Pitfall for Text-to-Image Generation

PosterCraft: Rethinking High-Quality Aesthetic Poster Generation in a Unified Framework

Reflective Flow Sampling Enhancement

Diffusion Dataset Condensation: Training Your Diffusion Model Faster with Less Data

Principal Spectral Regularization Makes Momentum Surpass Adam for LLM Training

IV-mixed Sampler: Leveraging Image Diffusion Models for Enhanced Video Synthesis

Zigzag Diffusion Sampling: Diffusion Models Can Self-Improve via Self-Reflection

OD³: Optimization-free Dataset Distillation for Object Detection

Noise Prompt Learning: Learning the Winning Tickets for Diffusion Sampling

The Blessing of Smooth Initialization for Video Diffusion Models