Guanjie Huang

Guangzhou, China

ghuang565@connect.hkust-gz.edu.cn

I am a Ph.D. candidate in Artificial Intelligence at The Hong Kong University of Science and Technology (Guangzhou), advised by Prof. Li Liu and Prof. Danny H. K. Tsang. My research focuses on efficient and reliable multimodal intelligence, especially audio-visual understanding, cued speech recognition, multimodal large language models, and uncertainty estimation.

My recent work studies how language models and specialized agents can reason over visual speech and hand cues, how multimodal systems can learn from limited data, and how their confidence and authenticity can be assessed. I am also interested in audio deepfake detection and physically grounded evaluation of audio-visual generation.

Before my doctoral study, I worked on industrial computer vision, large-scale multimodal dataset construction, UAV tracking, and learning-assisted optimization. I received an M.A.I. from the Australian National University and a B.Eng. in Software Engineering from the University of Electronic Science and Technology of China.

Research interests: audio-visual learning · multimodal large language models and agents · speech and cued speech recognition · trustworthy multimodal AI · model uncertainty · audio deepfake detection

I welcome conversations about research collaboration and multimodal AI opportunities.

news

May 06, 2026	Presented STF-ACSR, our semi training-free cued speech recognition work, at ICASSP 2026. Paper · Code
Jan 14, 2026	Our paper Semantic Modulated Prompting for Few-Shot Audio-Visual Classification was published in IEEE/ACM TASLP. Paper
Aug 01, 2025	Cued-Agent was selected for an oral presentation at ACM Multimedia 2025, together with a Student Travel Award. Paper · Code

selected publications

TASLP

Semantic Modulated Prompting for Few-Shot Audio-Visual Classification

Guanjie Huang, Yawen Cui, Danny Hin Kwok Tsang, and 2 more authors

IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2026

Audio-Visual LearningFew-Shot LearningPrompt TuningModality Rebalancing

DOI Bib HTML Code

@article{huang2026smp,
  title = {Semantic Modulated Prompting for Few-Shot Audio-Visual Classification},
  author = {Huang, Guanjie and Cui, Yawen and Tsang, Danny Hin Kwok and Wang, Wenwu and Liu, Li},
  journal = {IEEE/ACM Transactions on Audio, Speech, and Language Processing},
  volume = {34},
  pages = {723--736},
  year = {2026},
  doi = {10.1109/TASLPRO.2026.3654246},
}

ICASSP

Lend a Hand: Semi Training-Free Cued Speech Recognition via MLLM-Driven Hand Modeling

Guanjie Huang, Danny H. K. Tsang, Xiao-Ping Zhang, and 1 more author

In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2026

Cued Speech RecognitionMultimodal LLMTraining-Free LearningVisual Speech

Bib HTML Code

@inproceedings{huang2026lend,
  title = {Lend a Hand: Semi Training-Free Cued Speech Recognition via MLLM-Driven Hand Modeling},
  author = {Huang, Guanjie and Tsang, Danny H. K. and Zhang, Xiao-Ping and Liu, Li},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
  year = {2026},
}

ACM MM
Cued-Agent: A Multi-Agent Framework for Automatic Cued Speech Recognition

Guanjie Huang, Danny H. K. Tsang, Shan Yang, and 2 more authors

In Proceedings of the ACM International Conference on MultimediaOral presentation, 2025

Cued Speech RecognitionMulti-Agent SystemsMultimodal LearningLLM Self-Correction
Student Travel Award Bib HTML Code

Recipient of an ACM Multimedia 2025 Student Travel Award.
@inproceedings{huang2025cuedagent, title = {Cued-Agent: A Multi-Agent Framework for Automatic Cued Speech Recognition}, author = {Huang, Guanjie and Tsang, Danny H. K. and Yang, Shan and Lei, Guangzhi and Liu, Li}, booktitle = {Proceedings of the ACM International Conference on Multimedia}, year = {2025}, }
ACM MM
PhyAVBench: A Physical-Centric Audio-Visual Benchmark for Video Generation

Tianxin Xie, Wentao Lei, Guanjie Huang, and 1 more author

In Proceedings of the ACM International Conference on MultimediaOral presentation, 2026

Audio-Visual GenerationPhysical GroundingEvaluation BenchmarkText-to-Audio-Video
Bib HTML Website
@inproceedings{xie2026phyavbench, title = {PhyAVBench: A Physical-Centric Audio-Visual Benchmark for Video Generation}, author = {Xie, Tianxin and Lei, Wentao and Huang, Guanjie and others}, booktitle = {Proceedings of the ACM International Conference on Multimedia}, year = {2026}, }

TPAMI

WebUAV-3M: A Benchmark Unveiling the Power of Million-Scale Deep UAV Tracking

Chunhui Zhang, Guanjie Huang, Li Liu, and 5 more authors

IEEE Transactions on Pattern Analysis and Machine IntelligenceCo-first author. , 2023

UAV TrackingMultimodal TrackingLarge-Scale DatasetVisual Object Tracking

Bib HTML Code

@article{zhang2023webuav,
  title = {WebUAV-3M: A Benchmark Unveiling the Power of Million-Scale Deep UAV Tracking},
  author = {Zhang, Chunhui and Huang, Guanjie and Liu, Li and Huang, Shan and Yang, Yinan and Wan, Xiang and Ge, Shiming and Tao, Dacheng},
  journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year = {2023},
}

NeurIPS

WebUOT-1M: Advancing Deep Underwater Object Tracking with a Million-Scale Benchmark

Chunhui Zhang, Li Liu, Guanjie Huang, and 3 more authors

In Advances in Neural Information Processing Systems, 2024

Underwater TrackingLarge-Scale BenchmarkMultimodal TrackingObject Tracking

Bib HTML

@inproceedings{zhang2024webuot,
  title = {WebUOT-1M: Advancing Deep Underwater Object Tracking with a Million-Scale Benchmark},
  author = {Zhang, Chunhui and Liu, Li and Huang, Guanjie and Wen, Hao and Zhou, Xi and Wang, Yanfeng},
  booktitle = {Advances in Neural Information Processing Systems},
  year = {2024},
}

CVPRW

Underwater Camouflaged Object Tracking Meets Vision-Language SAM2

Chunhui Zhang, Li Liu, Guanjie Huang, and 5 more authors

In IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2025

Underwater TrackingCamouflaged ObjectsVision-Language ModelsSAM2

Outstanding Paper Bib HTML Code

Outstanding Paper Award at the CV4Animals Workshop.

@inproceedings{zhang2025underwater,
  title = {Underwater Camouflaged Object Tracking Meets Vision-Language SAM2},
  author = {Zhang, Chunhui and Liu, Li and Huang, Guanjie and Zhang, Zhipeng and Wen, Hao and Zhou, Xi and Ge, Shiming and Wang, Yanfeng},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
  year = {2025},
}

Google Scholar citations: 448