Fangyi Chen
Research Scientist
Intelligent Creation, ByteDance
Unsplashed background img 1
About Me

I am a Research Scientist in Generative AI at Intelligent Creation, ByteDance. My work focuses on post-training of multimodal large language model, with an emphasis on scene/video understanding and AI agent.

I received my Ph.D. in Electrical and Computer Engineering from Carnegie Mellon University in 2025, where I was affiliated with the CyLab Security & Privacy Institute and advised by Prof. Marios Savvides. Prior to that, I obtained my M.S. from the University of Pittsburgh in 2018, advised by Prof. Zhi-Hong Mao, and my B.E. from North China Electric Power University in 2017.

Publications
Referring Layer Decomposition
Fangyi Chen, Yaojie Shen, Lu Xu, Ye Yuan, Shu Zhang, Yulei Niu and Longyin Wen
International Conference on Learning Representations (ICLR), 2026
paper
code
MetaVLA: Unified Meta Co-training For Efficient Embodied Adaption
Chen Li, Zhantao Yang, Han Zhang, Fangyi Chen, Chenchen Zhu, Anudeep Bolimera and Marios Savvides
International Conference on Learning Representations (ICLR), 2026
paper
code
STELAR-VISION: Self-Topology-Aware Efficient Learning for Aligned Reasoning in Vision
Chen Li, Han Zhang, Zhantao Yang, Fangyi Chen, Zihan Wang, Anudeep Bolimera and Marios Savvides
AAAI Conference on Artificial Intelligence (AAAI), 2026
paper
code
Masked Autoencoders Are Effective Tokenizers for Diffusion Models
Hao Chen, Yujin Han, Fangyi Chen, Xiang Li, Yidong Wang, Jindong Wang, Ze Wang, Zicheng Liu, Difan Zou and Bhiksha Raj
International Conference on Machine Learning (ICML), 2025
paper
code
SoftVQ-VAE: Efficient 1-Dimensional Continuous Tokenizer
Hao Chen, Ze Wang, Xiang Li, Ximeng Sun, Fangyi Chen, Jiang Liu, Jindong Wang, Bhiksha Raj, Zicheng Liu and Emad Barsoum
Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2025
paper
code
RTGen: Generating Region-Text Pairs for Open-Vocabulary Object Detection
Fangyi Chen, Han Zhang, Zhantao Yang, Hao Chen, Kai Hu and Marios Savvides
Preprint
paper
code
Enhanced Training of Query-Based Object Detection via Selective Query Recollection
Fangyi Chen, Han Zhang, Kai Hu, Yukai Huang, Chenchen Zhu and Marios Savvides
Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR), 2023
paper
code
Unitail: Detecting, Reading, and Matching in Retail Scene
Fangyi Chen, Han Zhang, Zaiwang Li, Jiachen Dou, Shentong Mo, Hao Chen, Yongxin Zhang, Uzair Ahmed, Chenchen Zhu and Marios Savvides
European conference on computer vision (ECCV), 2022
paper
project website
code
Semantic Relation Reasoning for Shot-Stable Few-Shot Object Detection
Chenchen Zhu, Fangyi Chen, Uzair Ahmed, Zhiqiang Shen and Marios Savvides
Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR), 2021
paper
Soft Anchor-Point Object Detection
Chenchen Zhu, Fangyi Chen, Zhiqiang Shen and Marios Savvides
European conference on computer vision (ECCV), 2020
paper
NCMS: Towards accurate anchor free object detection through 𝓁2 norm calibration and multi-feature selection
Fangyi Chen, Chenchen Zhu, Zhiqiang Shen, Han Zhang and Marios Savvides
Computer Vision and Image Understanding (CVIU), Volumn 200, 103050
paper
Solving missing-annotation object detection with background recalibration loss
Han Zhang, Fangyi Chen, Zhiqiang Shen, Qiqi Hao, Chenchen Zhu and Marios Savvides
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2020
paper
code
Patents
Academic Services
    Conference Reviewer:
  •   • CVPR   • ICCV   • ECCV   • NeurIPS   • ICLR   • ICML
    Journal Reviewer:
  •   • TIP   • IJCV   • PR   • TGRS   • NeuralComputing
Teaching
    Graduate Teaching Assistant @ Pattern Recognition Theory, CMU, 2022 Fall
    Graduate Teaching Assistant @ Deep Learning Based Computer Vision and Pattern Recognition, CMU, 2023 Fall, 2024 Fall