Chieh-Hsin Lai

CMT: Mid-Training for Efficient Learning of Consistency, Mean Flow, and Flow-Map Models

Improved Object-Centric Diffusion Learning with Registers and Contrastive Alignment

SONA: Learning Conditional, Unconditional, and Matching-Aware Discriminator

Concept-TRAK: Understanding how diffusion models learn concepts through concept attribution

Improving Classifier-Free Guidance in Masked Diffusion: Low-Dim Theoretical Insights with High-Dim Impact

Weighted Point Set Embedding for Multimodal Contrastive Learning Toward Optimal Similarity Metric

SoundCTM: Unifying Score-based and Consistency Models for Full-band Text-to-Sound Generation

Jump Your Steps: Optimizing Sampling Schedule of Discrete Diffusion Models

HERO: Human-Feedback Efficient Reinforcement Learning for Online Diffusion Model Finetuning

G2D2: Gradient-guided Discrete Diffusion for image inverse problem solving

Bellman Diffusion: Generative Modeling as Learning a Linear Operator in the Distribution Space

VCT: Training Consistency Models with Variational Noise Coupling

Mitigating Embedding Collapse in Diffusion Models for Categorical Data