Ji Zhang

A new approach improves efficiency in Vision-Language Pre-training tasks.

A new method enhances OOD detection by combining global and local data representations.

A new approach improves task performance in vision-language models.

A new approach using multi-agent systems to enhance smaller language models.

This article discusses a new framework for assessing hallucinations in LVLMs.

A new benchmark evaluates how role-playing agents interact socially.

A new framework improves how language agents learn and perform tasks.

MIBench tests multimodal models' performance on multiple images.

mPLUG-Owl3 improves understanding of images and videos for better responses.

A new method to combine language models more effectively.

New modeling techniques enhance our understanding of bacterial movement.

MaVEn enhances AI's ability to process multiple images for better reasoning.

A new framework seeks to improve image generation using human feedback.

A look at how social media shapes collective opinions.

Discover how skip tuning enhances efficiency in vision-language models.