About me
👋 Hi, I’m Yu-Kai Huang, a PhD candidate in the Electrical and Computer Engineering Department at Carnegie Mellon University. I’m deeply passionate about connecting theoretical research with real-world applications in Computer Vision, Generative Models, and Autonomous Systems. My research explores topics such as generative models, trajectory prediction, and motion planning, all aimed at enabling intelligent systems to perceive, reason, and act effectively in complex environments.
Beyond the lab, I serve as a Research Team Lead in the Carnegie Mellon Blockchain Club, where I explore decentralized technologies that challenge conventional systems and open new possibilities.
I’m always excited to connect with fellow researchers, engineers, and innovators. Let’s build something meaningful together.
Awards
- Second and Third prizes in the 2025 ETHDenver BUIDLathon - 2025 ETHDenver BUIDLathon
- Second and Third places in the 2025 The Eigen Games - 2025 The Eigen Games
- First place in the 2025 TartanHacks: Story Prize - 2025 TartanHacks: Story Prize
- First place in the 2024 Waymo Open Dataset Challenges: Sim Agents Challenge - CVPR 2024 Workshop on Autonomous Driving
- Second place in the 2024 Waymo Open Dataset Challenges: Motion Prediction Challenge - CVPR 2024 Workshop on Autonomous Driving
- Champion of the 2024 Argoverse Multi-World Motion Forecasting Challenge - CVPR 2024 Workshop on Autonomous Driving
- Champion of the 2023 Argoverse Motion Forecasting Challenge - CVPR 2023 Workshop on Autonomous Driving
- Rank 1 in the Argoverse 2: Multi-Agent Motion Forecasting Competition - 2023 Argoverse 2: Multi-Agent Motion Forecasting Competition
- Rank 1 in the Argoverse 2: Single-Agent Motion Forecasting Competition - 2023 Argoverse 2: Single-Agent Motion Forecasting Competition
- Rank 1 in the Argoverse 1: Single-Agent Motion Forecasting Competition - 2023 Argoverse 1: Single-Agent Motion Forecasting Competition
- Rank 5 in the MegaFace Challenge 1: Unrestricted recognition with varying number of distractors - 2018 MegaFace Challenge 1
Publications
- VTutor: An Animated Pedagogical Agent SDK that Provide Real Time Multi-Model Feedback - AIED, 2025
- ModeSeq: Taming Sparse Multimodal Motion Prediction with Sequential Mode Modeling - CVPR, 2025
- BehaviorGPT: Smart Agent Simulation for Autonomous Driving with Next-Patch Prediction - NeurIPS, 2024
- A Reference-Based 3D Semantic-Aware Framework for Accurate Local Facial Attribute Editing - IEEE International Joint Conference on Biometrics (IJCB), 2024
- TOFG: Temporal Occupancy Flow Graph for Prediction and Planning in Autonomous Driving - IEEE Transactions on Intelligent Vehicles (TIV), 2023
- QCNeXt: A Next-Generation Framework For Joint Multi-Agent Trajectory Prediction - arXiv, 2023
- Enhanced Training of Query-Based Object Detection via Selective Query Recollection - CVPR, 2023
- Query-Centric Trajectory Prediction - CVPR, 2023
- Unsupervised Disentanglement of Linear-Encoded Facial Semantics - CVPR, 2021