Wansen Wu

National University of Defense Technology

Papers

Total Citations

H-Index

About

Wansen Wu is a researcher advancing the frontier of embodied AI, with a primary focus on **Vision-Language Navigation (VLN)** —a field that bridges natural language processing, computer vision, and robotics. Wu’s most significant contribution is a comprehensive **survey and taxonomy of VLN** (2023, 38 citations), which systematically categorizes the diverse approaches to training agents that follow human language instructions to navigate unfamiliar, real-world environments. This work has become a key reference for researchers tackling the challenge of long-horizon planning in embodied agents. Building on this foundation, Wu proposed a **Self-Organizing Memory mechanism based on Adaptive Resonance Theory** (2023), a novel architecture designed to improve an agent’s ability to plan over extended sequences by dynamically storing and retrieving spatial-semantic knowledge. This work directly addresses a critical bottleneck in VLN: the need for robust, long-term memory in complex environments. With a growing citation footprint, Wansen Wu is establishing a reputation for both synthesizing the field’s progress and introducing principled, biologically inspired solutions to its hardest problems.