Flood Sung
Papers
1
Total Citations
99
H-Index
1
About
Flood Sung is a prominent researcher in artificial intelligence, with a primary focus on computer vision and natural language processing, particularly at the intersection of visual intelligence and human communication. His most cited work, "Actor-Critic Sequence Training for Image Captioning" (2017, 99 citations), introduced a novel reinforcement learning approach to generate natural language descriptions of images—a critical capability for AI agents that must interact with human users about their visual surroundings. By applying actor-critic methods to sequence generation, Sung addressed the limitations of traditional likelihood-based training, enabling more coherent and contextually relevant captions. This contribution has been instrumental in advancing how robots and visual-intelligence systems perceive and describe the world, bridging the gap between raw visual data and human-readable language. Sung's work stands out for its practical impact on human-AI interaction, demonstrating how reinforcement learning can refine complex generative tasks. His research continues to influence the development of more communicative and perceptive AI agents, making him a key figure in the evolution of vision-language models.
Research Focus
Key Achievements
Top Papers
- 1Actor-Critic Sequence Training for Image Captioning99 citations · 2017