A Life-long Student.
An Innocent Guy.
Hey, I am Yifan Chen, a Stats PhD student at UIUC, advised by Prof. Yun Yang.
My research interest lies in the general area of computation in machine learning, currently in the application of computational statistics to neural network acceleration. I focus on the models with heavy matrix operation, including Transformers and graph neural networks (GNNs).
I obtained my B.S. degree from Department of Statistics, Fudan University in 2018, advised by Prof. Juan Shen and Prof. Chenghong Zhang.
Publication
Inducer-tuning: Connecting Prefix-tuning and Adapter-tuning
Yifan Chen*, Devamanyu Hazarika*, Mahdi Namazifar, Yang Liu, Di Jin, Dilek Hakkani-Tur
The 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP 2022)Sketching as a Tool for Understanding and Accelerating Self-attention for Long Sequences
Yifan Chen*, Qi Zeng*, Dilek Hakkani-Tur, Di Jin, Heng Ji, Yun Yang
2022 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2022) OralEmpowering parameter-efficient transfer learning by recognizing the kernel structure in self-attention
Yifan Chen*, Devamanyu Hazarika*, Mahdi Namazifar, Yang Liu, Di Jin, Dilek Hakkani-Tur
2022 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2022) FindingsSkyformer: Remodel Self-Attention with Gaussian Kernel and Nyström Method
Yifan Chen*, Qi Zeng*, Heng Ji, Yun Yang
Thirty-Fifth Conference on Neural Information Processing Systems (NeurIPS) 2021Accumulations of Projections—A Unified Framework for Random Sketches in Kernel Ridge Regression
Yifan Chen, Yun Yang
International Conference on Artificial Intelligence and Statistics (AISTATS) 2021Fast Statistical Leverage Score Approximation in Kernel Ridge Regression
Yifan Chen, Yun Yang
International Conference on Artificial Intelligence and Statistics (AISTATS) 2021Narrowing the Gap between Professionality and Popularity: An Empirical Investigation on Community QA
Chenghong Zhang, Yifan Chen, Hongyue Lan, Yicheng Zhang, Tian Lu
American Conference on Information Systems (AMCIS) Session on TREO 2017
