Yuhang Song

Office A1.09

Ashton Building, Brownlow street

Liverpool, L3 3GQ

I am a Ph.D. student at the University of Liverpool and National Tsing Hua University, working at the intersection of embodied intelligence, deep learning, and autonomous systems.

My research life moves between airvLab at the Digital Innovation Facility and ELSA Lab at NTHU. I founded TOPSOFT Group in 2016 as an international organization for computer science.

I am interested in agents that can perceive, decide, and act with efficiency in the physical world. The long-term aim is to build robust end-to-end navigation systems that feel less like isolated models and more like reliable companions for real environments.

Deep Learning Deep Reinforcement Learning Embodied AI Autonomous Systems Efficient Robotics

News

Apr 1, 2026	Two papers accepted to ICPR 2026 !
Jul 14, 2025	Started AMD full-time internship in Shanghai! Focusing on Embodied AI with rocm.
Oct 12, 2023	I serverd as external reviewer at ICLR 2023.
Sep 11, 2023	One paper is accepted to ACML 2023 !
Aug 14, 2023	Building Real World Aligned Embodied Autonomous Navigation System! <3
Jun 20, 2023	Started my journey on learning ROS 1 & 2, preparing working on embodied Husky robots.
Jan 18, 2023	I serve as module demonstrator for COM 341 in UoL.
Oct 1, 2022	Learning on RL basics and Efficient Models.
Jul 19, 2022	I got an fully-funded PhD offer from University of LIverpool.
Jun 18, 2022	Received my Bachelor degree from University of Liverpool.

Latest Posts

Oct 16, 2025	Running OpenVLA Evaluation (LIBERO) on AMD Servers
Jul 2, 2023	Installation Updates of Sim-to-Real-Virtual-Guidance-for-Robot-Navigation
Nov 8, 2022	Advanced Reinforcement Learning
Oct 30, 2022	Reinforcement Learning Basics
Oct 28, 2022	Study on Efficient Models 2
Oct 7, 2022	Study on Efficient Models

Publications

2026

ICPR

Fine-Grained Alignment in Vision-and-Language Navigation through Bayesian Optimization

Yuhang Song, Mario Gianni, Chenguang Yang, Kunyang Lin, Te-Chuan Chiu, Anh Nguyen, and Chun-Yi Lee

International Conference on Pattern Recognition, 2026

ICPR 2026

Bib

@article{song2024fine,
  title = {Fine-Grained Alignment in Vision-and-Language Navigation through Bayesian Optimization},
  author = {Song, Yuhang and Gianni, Mario and Yang, Chenguang and Lin, Kunyang and Chiu, Te-Chuan and Nguyen, Anh and Lee, Chun-Yi},
  journal = {International Conference on Pattern Recognition},
  year = {2026}
}

ICPR

Dynamic Context Adapters: Efficiently Infusing History into Vision-and-Language Models

Yuhang Song, Bor-Jiun Lin, Jiaxu Liu, Te-Chuan Chiu, Anh Nguyen, and Chun-Yi Lee

International Conference on Pattern Recognition, 2026

ICPR 2026

Bib

@article{song2026dynamic,
  title = {Dynamic Context Adapters: Efficiently Infusing History into Vision-and-Language Models},
  author = {Song, Yuhang and Lin, Bor-Jiun and Liu, Jiaxu and Chiu, Te-Chuan and Nguyen, Anh and Lee, Chun-Yi},
  journal = {International Conference on Pattern Recognition},
  year = {2026}
}

2025

TMLR

Toward Linearly Regularizing the Geometric Bottleneck of Linear Generalized Attention

Jiaxu Liu, Xinping Yi, Xiangyu Yin, Yuhang Song, Gaojie Jin, and Xiaowei Huang

Transactions on Machine Learning Research, 2025

TMLR 2025

Bib

@article{liu2025toward,
  title = {Toward Linearly Regularizing the Geometric Bottleneck of Linear Generalized Attention},
  author = {Liu, Jiaxu and Yi, Xinping and Yin, Xiangyu and Song, Yuhang and Jin, Gaojie and Huang, Xiaowei},
  journal = {Transactions on Machine Learning Research},
  year = {2025}
}

2024

Learning to Terminate in Object Navigation

Yuhang Song, Anh Nguyen, and Chun-Yi Lee

In Asian Conference on Machine Learning, 2024

ACML 2024

Bib HTML Code

@inproceedings{song2024learning,
  title = {Learning to Terminate in Object Navigation},
  author = {Song, Yuhang and Nguyen, Anh and Lee, Chun-Yi},
  booktitle = {Asian Conference on Machine Learning},
  pages = {1247--1262},
  year = {2024},
  organization = {PMLR},
}