Li Zhang

Northumbria University, Harbin Institute of Technology

Papers

Total Citations

108

H-Index

About

Li Zhang is a researcher whose work spans the intersection of artificial intelligence, computer vision, and robotics, with a particular focus on vision-language understanding and autonomous systems. Zhang's most influential contribution, "Actor-Critic Sequence Training for Image Captioning" (2017), addresses a fundamental challenge in visual AI: enabling machines to generate meaningful natural language descriptions of visual scenes — a capability critical for robots communicating with human users. This work, which applies reinforcement learning techniques to optimize sequence generation beyond traditional likelihood-based training, has garnered 99 citations, reflecting its significant uptake within the computer vision and natural language processing communities. Complementing this, Zhang's work on semi-supervised vision-language mapping through variational learning tackles the persistent problem of limited labeled training data in cross-modal retrieval tasks, bridging image and sentence understanding in low-resource settings. An earlier contribution to agent-based modeling for remote robotic welding systems demonstrates Zhang's longer-standing interest in intelligent robotic control. Collectively, Zhang's research traces a compelling trajectory from applied robotics toward sophisticated AI systems capable of perceiving, reasoning about, and communicating the visual world in human-interpretable language.