Sinan Tan

Papers

Total Citations

H-Index

About

Sinan Tan is a researcher at the forefront of embodied AI and human-robot collaboration, with a focus on bridging the gap between natural language instructions and autonomous robotic action. His work centers on two key challenges: enabling robots to interpret ambiguous human commands and developing self-supervised learning methods for vision-and-language navigation. In his highly cited 2022 paper, "Embodied Multi-Agent Task Planning from Ambiguous Instruction" (24 citations), Tan tackles the critical problem of implicit information in human-robot communication, proposing frameworks that allow multiple agents to collaboratively infer and execute tasks from incomplete or vague instructions—a fundamental step toward intuitive human-robot teamwork. Complementing this, his work on "Self-supervised 3D Semantic Representation Learning for Vision-and-Language Navigation" (6 citations) advances the ability of embodied agents to navigate complex environments by learning rich 3D semantic representations without costly manual annotations. This research is pivotal for practical applications in service robotics and autonomous systems. Tan’s contributions are shaping how robots understand and act upon human language in the real world, making him a rising voice in the intersection of computer vision, NLP, and robotics.