Hi! I am Jiaqi Xiong (ηεηͺ). I am a Master student in Social Data Science at the Oxford Internet Institute, University of Oxford, where I am fortunate to be supervised by Prof. Xiaowen Dong and Prof. Michael Bronstein, and to work closely with Dr. Andy Huang. Before Oxford, I obtained my dual bachelorβs degree in Artificial Intelligence from the University of Aberdeen (with first-class honours) and South China Normal University (SCNU) in 2025, where I completed my thesis under the supervision of Dr. Chris Norval. My research focuses on applying graph neural networks (GNNs) to biological / social sciences and sports analytics, and on building trustworthy agentic AI systems and multimodal AI.
π₯ News
-
One paper accepted to CVPR 2026.
-
One paper accepted to NeurIPS 2025 workshop.
-
Joined the University of Oxford as a master student.
-
One paper accepted to BMC Bioinformatics.
-
Obtained bachelorβs degree from University of Aberdeen and South China Normal University.
-
One paper accepted to the 33rd Conference on Intelligent Systems for Molecular Biology (ISMB).
π Education
-
MSc, Social Data Science
University of Oxford
-
BSc, Artificial Intelligence
University of Aberdeen
First-class Honours -
BEng, Artificial Intelligence
South China Normal University
Outstanding Graduate
π Publications
β
First author
DEAF: A Benchmark for Diagnostic Evaluation of Acoustic Faithfulness in Audio Language Models
@article{xiong2026deaf,
title = {{DEAF}: A Benchmark for Diagnostic Evaluation of Acoustic Faithfulness in Audio Language Models},
author = {Xiong, Jiaqi and Qi, Yunjia and Cao, Qi and Zheng, Yu and Zhang, Yutong and Wang, Ziteng and Liao, Ruofan and Xu, Weisheng and Liu, Sichen},
year = {2026},
eprint = {2603.18048},
archivePrefix = {arXiv},
primaryClass = {cs.AI}
}
Iterative Closed-Loop Motion Synthesis for Scaling the Capabilities of Humanoid Control
@inproceedings{xu2026iterative,
title = {Iterative Closed-Loop Motion Synthesis for Scaling the Capabilities of Humanoid Control},
author = {Xu, Weisheng and Wu, Qiwei and Zhang, Jiaxi and Jing, Tan and Li, Yangfan and Fang, Yuetong and Xiong, Jiaqi and Wu, Kai and Ou, Rong and Xu, Renjing},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
year = {2026},
eprint = {2602.21599},
archivePrefix = {arXiv},
primaryClass = {cs.RO}
}
V-FAT: Benchmarking Visual Fidelity Against Text-bias
@article{wang2026vfat,
title = {{V-FAT}: Benchmarking Visual Fidelity Against Text-bias},
author = {Wang, Ziteng and He, Yujie and Li, Guanliang and Yang, Siqi and Xiong, Jiaqi and Liu, Songxiang},
year = {2026},
eprint = {2601.04897},
archivePrefix = {arXiv},
primaryClass = {cs.CL}
}
β
First author
Psychometrically Grounded Evaluation of LLM Personas
@inproceedings{xiong2025psychometrically,
title = {Psychometrically Grounded Evaluation of {LLM} Personas},
author = {Xiong, Jiaqi},
booktitle = {First Workshop on {LLM} Persona Modeling at the Conference on Neural Information Processing Systems (NeurIPS)},
year = {2025},
url = {https://openreview.net/forum?id=j1jDozwwmx}
}
β
First author
Cross-Attention Graph Neural Networks for Inferring Gene Regulatory Networks with Skewed Degree Distribution
@article{xiong2025crossattention,
title = {Cross-Attention Graph Neural Networks for Inferring Gene Regulatory Networks with Skewed Degree Distribution},
author = {Xiong, Jiaqi and Yin, Nan and Liang, Shiyang and Li, Haoyang and Wang, Yingxu and Ai, Duo and Wang, Jingjie},
journal = {BMC Bioinformatics},
volume = {26},
number = {1},
year = {2025},
publisher = {Springer Nature},
doi = {10.1186/s12859-025-06186-1}
}
MutBERT: Probabilistic Genome Representation Improves Genomics Foundation Models
@article{long2025mutbert,
title = {{MutBERT}: Probabilistic Genome Representation Improves Genomics Foundation Models},
author = {Long, Weicai and Su, Houcheng and Xiong, Jiaqi and Zhang, Yanlin},
journal = {bioRxiv},
year = {2025},
doi = {10.1101/2025.01.23.634452},
note = {Accepted at ISMB 2025}
}
PillarTrack: Boosting Pillar Representation for Transformer-based 3D Single Object Tracking on Point Clouds
@article{xu2024pillartrack,
title = {{PillarTrack}: Boosting Pillar Representation for Transformer-based {3D} Single Object Tracking on Point Clouds},
author = {Xu, Weisheng and Zhou, Sifan and Xiong, Jiaqi and Zhao, Ziyu and Yuan, Zhihang},
year = {2024},
eprint = {2404.07495},
archivePrefix = {arXiv},
primaryClass = {cs.CV}
}π Honors and Awards
π Honors and Scholarship
-
2025.06
Honorary Scholarship for Undergraduate Graduates
Aberdeen Institute of South China Normal University
-
2025.03
First Prize, Innovation Award Scholarship
South China Normal University
-
2025.01
First Prize, Outstanding Student Scholarship
South China Normal University
π Competition Awards
-
2025.04
Silver Prize
China International College Students' Innovation Competition 2025
-
2024.08
Third Prize
National Final of China Robotics and Artificial Intelligence Competition
-
2024.07
First Prize
Guangdong College Computer Design Competition
-
2024.07
First Prize
Network Technology Challenge of China Collegiate Computing Competition
π¬ Contact
Reach me at jiaqi.xiong@oii.ox.ac.uk, or find me on GitHub Β· Google Scholar Β· LinkedIn.
Based in Oxford, UK. Open to research collaboration and chats.