Hao Chen

KnowledgeSmith: Uncovering Knowledge Updating in LLMs with Model Editing and Unlearning

ImageDoctor: Diagnosing Text-to-Image Generation via Grounded Image Reasoning

Turning Internal Gap into Self-Improvement: Promoting the Generation-Understanding Unification in MLLMs

TrustJudge: Inconsistencies of LLM-as-a-Judge and How to Alleviate Them

CAT-VIDEO: CORRUPTION-AWARE TRAINING FOR ROBUST VIDEO DIFFUSION MODELS

On the Diversity of Synthetic Data and its Impact on Training Large Language Models

Image Tokenizer Needs Post-Training

APRIL: Active Partial Rollouts in Reinforcement Learning to Tame Long-tail Generation

On alignment of unified multimodal large language models

From Pretraining to Pathology: How Noise Leads to Catastrophic Inheritance in Medical Models

Rethinking the Bias of Foundation Model under Long-tailed Distribution

On Fairness of Unified Multimodal Large Language Model for Image Generation

Unleashing Hour-Scale Video Training for Long Video-Language Understanding

Masked Autoencoders Are Effective Tokenizers for Diffusion Models

ImageFolder: Autoregressive Image Generation with Folded Tokens

Automatic Dataset Construction (ADC): Sample Collection, Data Curation, and Beyond

On the Diversity of Synthetic Data and its Impact on Training Large Language Models

ControlVAR: Exploring Controllable Visual Autoregressive Modeling

On the Robustness of Vision-Language Models Against Distractions