Siyuan Yang

Wallenberg–NTU Presidential Postdoctoral Fellow

Division of Robotics, Perception and Learning, KTH Royal Institute of Technology

I am a Wallenberg–NTU Presidential Postdoctoral Fellow, currently hosted at the Division of Robotics, Perception and Learning, KTH Royal Institute of Technology and working with Prof. Hedvig Kjellström. During the first year of my fellowship, I was based at the College of Computing and Data Science, Nanyang Technological University, where I worked with Prof. Weisi Lin.

My research focuses on human-centred visual intelligence, particularly human activity understanding, motion representation learning, and 3D human motion generation. I develop methods that learn with limited supervision and connect human motion with pretrained vision and generative models.

I received my Ph.D. and M.Sc. degrees from Nanyang Technological University. My Ph.D. research was conducted at the Rapid-Rich Object Search Lab and supervised by Prof. Alex C. Kot and Assoc. Prof. Shijian Lu. During my Ph.D., I also worked closely with Prof. Jun Liu. I received my B.Eng. degree from Harbin Institute of Technology.

siyuany@kth.se KTH Profile Google Scholar CV

Preprints

Image Diffusion Models Are Good 3D Human Motion Generators

Siyuan Yang*, Chong Wang*, Hao Cheng, Jun Liu, Bihan Wen, Alex C. Kot, Hedvig Kjellström

arXiv preprint, 2026 · arXiv:2610.14254

Skeleton-to-Image Encoding: Enabling Skeleton Representation Learning via Vision-Pretrained Models

Siyuan Yang, Jun Liu, Hao Cheng, Chong Wang, Shijian Lu, Hedvig Kjellström, Weisi Lin, Alex C. Kot

arXiv preprint, 2026 · arXiv:2603.05963

TAME: Topology-Aware Text-Driven Motion Editing across Heterogeneous Humanoid Skeletons

Qichen Zheng, Siyuan Yang^#, Chong Wang, Jun Liu, Shijian Lu, Alex C. Kot, Kwok-Yan Lam

arXiv preprint, 2026 · arXiv:2610.04529

Persistence Forcing: Exploiting Feature Specialization in Pixel-Space Diffusion

Chong Wang, Zixuan Fu, Shiqi Huang, Siyuan Yang, Hao Cheng, Bihan Wen

arXiv preprint, 2026 · arXiv:2609.36014

* Equal contribution. ^# Corresponding author.

Selected Publications

For a complete and up-to-date publication list, please see my Google Scholar profile.

One-Shot Action Recognition via Multi-Scale Spatial-Temporal Skeleton Matching

Siyuan Yang, Jun Liu, Shijian Lu, Meng Hwa Er, Alex C. Kot

IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024 · arXiv

Self-Supervised 3D Action Representation Learning with Skeleton Cloud Colorization

Siyuan Yang, Jun Liu, Shijian Lu, Meng Hwa Er, Yongjian Hu, Alex C. Kot

IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023 · arXiv

Skeleton Cloud Colorization for Unsupervised 3D Action Representation Learning

Siyuan Yang, Jun Liu, Shijian Lu, Meng Hwa Er, Alex C. Kot

ICCV, 2021 · arXiv

Collaborative Learning of Gesture Recognition and 3D Hand Pose Estimation with Multi-Order Feature Analysis Spotlight

Siyuan Yang, Jun Liu, Shijian Lu, Meng Hwa Er, Alex C. Kot

ECCV, 2020 · Paper

Towards Physical World Backdoor Attacks against Skeleton Action Recognition

Qichen Zheng, Yi Yu, Siyuan Yang^#, Jun Liu, Kwok-Yan Lam, Alex C. Kot

ECCV, 2024 · arXiv · Project

Benchmarking the Robustness of Spatial-Temporal Models Against Corruptions

Chenyu Yi*, Siyuan Yang*, Haoliang Li, Yap-Peng Tan, Alex C. Kot

NeurIPS Datasets and Benchmarks Track, 2021 · Paper · Code

Unlearnable Examples Detection via Iterative Filtering Oral

Yi Yu*, Qichen Zheng*, Siyuan Yang^#, Wenhan Yang, Jun Liu, Shijian Lu, Yap-Peng Tan, Kwok-Yan Lam, Alex C. Kot

ICANN, 2024 · arXiv

FaceComposer: Learning Coherent and Realistic Face Composition from References Oral

Ling Li, Lanqing Guo, Siyuan Yang, Yufei Wang, Qian Zheng, Alex C. Kot, Weisi Lin, Lihui Chen

IJCNN, 2026 · Model

Beyond the Horizon: Decoupling Multi-View UAV Action Recognition via Partial Order Transfer

Wenxuan Liu, Zhuo Zhou, Xuemei Jia, Siyuan Yang, Wenxin Huang, Xian Zhong, Chia-Wen Lin

AAAI, 2026 · Paper

Proactive Image Manipulation Detection and Tracing in Fake News

Ruohan Meng, Siyuan Yang, Zhili Zhou, Qi Cui, Kwok-Yan Lam, Zengwei Zheng, Alex C. Kot

IEEE Transactions on Dependable and Secure Computing, 2026 · DOI

Vid-Group: Temporal Video Grounding Pretraining from Unlabeled Videos in the Wild

Peijun Bao, Chenqi Kong, Siyuan Yang, Zihao Shao, Xinghao Jiang, Boon Poh Ng, Meng Hwa Er, Alex C. Kot

ICCV, 2025 · Paper · Code

MTL-UE: Learning to Learn Nothing for Multi-Task Learning

Yi Yu, Song Xia, Siyuan Yang, Chenqi Kong, Wenhan Yang, Shijian Lu, Yap-Peng Tan, Alex C. Kot

ICML, 2025 · Paper · Code

Reconciling Stochastic and Deterministic Strategies for Zero-shot Image Restoration using Diffusion Model in Dual

Chong Wang, Lanqing Guo, Zixuan Fu, Siyuan Yang, Hao Cheng, Alex C. Kot, Bihan Wen

CVPR, 2025 · arXiv · Code

STSP: Spatial-Temporal Subspace Projection for Video Class-Incremental Learning

Hao Cheng, Siyuan Yang, Chong Wang, Joey Tianyi Zhou, Alex C. Kot, Bihan Wen

ECCV, 2024 · Paper

Frequency Guidance Matters in Few-Shot Learning Oral

Hao Cheng, Siyuan Yang, Joey Tianyi Zhou, Lanqing Guo, Bihan Wen

ICCV, 2023 · Paper · Code

Temporal Coherent Test-Time Optimization for Robust Video Classification

Chenyu Yi, Siyuan Yang, Yufei Wang, Haoliang Li, Yap-Peng Tan, Alex C. Kot

ICLR, 2023 · arXiv

* Equal contribution. ^# Corresponding author.

Research Funding

Principal Investigator (PI): All-in-One LLM Framework for 3D Human Motion Understanding, Nanyang Technological University, SGD 100,000, 2024
Principal Investigator (PI): All-in-One LLM Framework for 3D Human Motion Understanding, Knut and Alice Wallenberg Foundation, SEK 300,000, 2026

Teaching

KTH Royal Institute of Technology

Guest Lecturer: DT2140: Multimodal Interaction and Interfaces (2026)

Nanyang Technological University

Lab Instructor: IE2110: Signals and Systems (2025)
Teaching Assistant: EE1005: C Programming; IE2110 & EE2010: Signals and Systems; EE4483: Artificial Intelligence and Data Mining

Professional Service

Area Chair:: ICLR 2027, ICML 2027, BMVC 2026, VCIP 2026
Tutorial Co-organizer:: ECCV 2026 — From Perception to Simulation: The Emergence of World Models in Multi-modal Reasoning
Tutorial Speaker:: ICPR 2026 — Advancing Comprehensive Reasoning in Multimodal Large Language Models
Registration Chair:: BMVC 2026
Journal Reviewer:: TPAMI, IJCV, TIP, TMM, TII, TCSVT, TNNLS, PR, and others
Conference Reviewer:: CVPR, ICCV, ECCV, NeurIPS, ICML, ICLR, AAAI, ACMMM, WACV, BMVC, ICME, ICASSP, and others

Awards

Wallenberg–NTU Presidential Postdoctoral Fellowship, 2024
Singapore Data Science Consortium Dissertation Research Fellowship, 2021 (awarded to 10 doctoral researchers annually)