Ph.D. Student · Research Intern · Embodied AI

Junli Liu

My research explores how embodied agents understand instructions, navigate complex environments, and build predictive world models.

I am a Ph.D. student at Northwestern Polytechnical University, where I also completed my bachelor's degree, and I am currently a research intern at Shanghai Artificial Intelligence Laboratory.

Email GitHub Scholar

Updates

News

  1. FreeGaussian was published in the AAAI 2026 proceedings.

  2. OpenFly was accepted to ICLR 2026.

  3. AerialVG was accepted to ICCV 2025.

Focus

What I Work On

Embodied AI

Building agents that perceive, reason, and act in physical or simulated environments.

Vision-Language Navigation

Connecting visual perception and language instructions for navigation and decision making.

World Model

Learning structured representations that help agents predict, plan, and generalize.

Selected Work

Publications and Projects

AerialVG teaser figure

ICCV 2025

AerialVG: A Challenging Benchmark for Aerial Visual Grounding by Exploring Positional Relations

Junli Liu, Qizhi Chen, Zhigang Wang, Yiwen Tang, Yiting Zhang, Chi Yan, Dong Wang, Xuelong Li, Bin Zhao

AerialVG introduces an aerial-view visual grounding benchmark that emphasizes positional relations and spatial reasoning in high-resolution remote imagery.

Paper arXiv Code
OpenFly cover figure

ICLR 2026

OpenFly: A Comprehensive Platform for Aerial Vision-Language Navigation

Yunpeng Gao, Chenhui Li, Zhongrui You, Junli Liu, Zhen Li, Pengan Chen, Qizhi Chen, et al.

OpenFly provides an automated toolchain, large-scale benchmark, and OpenFly-Agent for outdoor aerial vision-language navigation.

Project page Paper Code
FreeGaussian teaser figure

AAAI 2026 · Oral

FreeGaussian: Annotation-free Control of Articulated Objects via 3D Gaussian Splats with Flow Derivatives

Qizhi Chen, Delin Qu, Junli Liu, Yiwen Tang, Haoming Song, Dong Wang, Bin Zhao, et al.

FreeGaussian reconstructs controllable 3D Gaussian representations for articulated objects from monocular video without manual annotations.

Project page Paper Code

Background

Experience

Research Intern, Shanghai Artificial Intelligence Laboratory

Working on research topics around embodied intelligence, vision-language navigation, and world models.

Ph.D. Student, Northwestern Polytechnical University

Researching Embodied AI, Vision-Language Navigation, and world models.

Undergraduate Student, Northwestern Polytechnical University

Completed undergraduate study at Northwestern Polytechnical University.

Contact

Let's Connect

I am open to research conversations, collaboration, and practical projects around Embodied AI, Vision-Language Navigation, and world models.

liujunlideal@gmail.com github.com/Ideal-ljl Google Scholar