Jing Shi

Vision-Zero: Scalable VLM Self-Evolution via Multi-Agent Self-Play

SupGRPO: Enhancing GRPO with Matching-based Online SFT for Text Spotting

Seeing Through Words: Controlling Visual Retrieval Quality with Language Models

VividCam: Learning Unconventional Camera Motions from Virtual Synthetic Videos

Understanding and Mitigating Numerical Sources of Nondeterminism in LLM Inference

Toward Robust Hyper-Detailed Image Captioning: A Multiagent Approach and Dual Evaluation Metrics for Factuality and Coverage

GroundingBooth: Grounding Text-to-Image Customization

Fusing Visual and Textual Cues for Sequential Image Difference Captioning

DiffTell: A Comprehensive Dataset for Image Difference Captioning

MAGNET: Augmenting Generative Decoders with Representation Learning and Infilling Capabilities

Enhancing Factuality in Detailed Image Captioning with LLM-MLLM Collaboration