Hou Pong Chan

Papers

Total Citations

H-Index

About

Hou Pong Chan is a researcher advancing the intersection of multimedia understanding and task-oriented artificial intelligence. His primary research areas include generative script learning, task planning, and multimodal reasoning—fields that aim to equip AI systems with the ability to anticipate and execute goal-directed activities. Chan’s most notable contribution is his work on *Multimedia Generative Script Learning for Task Planning* (2023), which tackles the challenge of generating subsequent steps to achieve a specific goal by integrating visual historical states. This approach is critical for enabling robots and intelligent agents to perform everyday tasks by learning from both textual and visual cues, bridging the gap between static knowledge and dynamic, real-world execution. Although early in its trajectory, this work has already garnered 5 citations, signaling growing interest in his methodology. Chan’s research holds promise for applications in robotics, human-computer interaction, and automated planning, where understanding context and sequential logic is paramount. His focus on grounding generative models in multimodal data represents a forward-looking step toward more adaptive and perceptive AI systems.