🤗 Hello! I am Pengjun Fang. Nice to meet you!
👨‍💻 I am a MPhil student in Computer Science at HKUST, advised by Prof. Qifeng Chen.
👨‍💻 My research focuses on generative AI, in particular, multimodal generation, video-to-audio synthesis, and controllable video generation.
📫 How to reach me: pfangaf@connect.ust.hk

More recent projects:

• [ICLR 2026] AC-Foley: Reference-audio-guided video-to-audio synthesis with acoustic transfer. [paper] [project]

• AnimationBench: A benchmark for evaluating character-centric animation capabilities of video models. [paper] [project]

• WanFM: First-Last-Frame-to-Video generation with bidirectional denoising and last-frame constraints.

• VideoTuna: An open-source codebase for text-to-video, image-to-video, and text-to-image generation.

For more of my generative AI projects and publications, please check my personal webpage and Google Scholar.