Hi, There! This is Yihua Cheng, a 4th-year CS Ph.D. in University of Chicago, advised by Junchen Jiang . My research focuses on computer networks and systems and I’m interested in systems for data streaming. Specifically, my research covers the real-time video streaming, streaming data analysis, and KV cache streaming optimization for LLM serving. I got my bachelor degree in Peking University in 2020. Here is my CV.
Email: yihua98@uchicago.edu
Github: https://github.com/ApostaC
Recent publications
GRACE: Loss-Resilient Real-Time Video through Neural Codecs
Yihua Cheng, Ziyi Zhang, Hanchen Li, Anton Arapin, Yue Zhang, Yuhan Liu, Kuntai Du, Xu Zhang, Francis Y. Yan, Amrita Mazumdar, Nick Feamster, Junchen Jiang
NSDI 2024
Project website, Demo video (coming soon), Talk slides
CacheGen: KV Cache Compression and Streaming for Fast Language Model Serving
Yuhan Liu, Hanchen Li, Yihua Cheng, Siddhant Ray, Yuyang Huang, Qizheng Zhang, Kuntai Du, Jiayi Yao, Shan Lu, Ganesh Ananthanarayanan, Michael Maire, Henry Hoffmann, Ari Holtzman, Junchen Jiang
In submission
Earth+: on-board satellite imagery compression leveraging historical earth observations
Kuntai Du, Yihua Cheng, Peder Olsen, Shadi Noghabi, Ranveer Chandra, Junchen Jiang
In submission
Online Profiling and Adaptation of Quality Sensitivity for Internet Video
Yihua Cheng, Hui Zhang, Junchen Jiang
SoCC 2023
Raising the Level of Abstraction for Time-State Analytics With the Timeline Framework
Henry Milner, Yihua Cheng, Jibin Zhan, Hui Zhang, Vyas Sekar, Junchen Jiang, Ion Stoica
CIDR 2023
Enabling Perception-Driven Optimization in Networking
Yihua Cheng, Xu Zhang, Junchen Jiang
SIGMETRICS 2023
Yunzhe Ni, Feng Qian, Taide Liu, Yihua Cheng, Zhiyao Ma, Jing Wang, Zhongfeng Wang, Gang Huang, Xuanzhe Liu, Chenren Xu
NSDI 2023
An Active-Passive Measurement Study of TCP Performance over LTE on High-speed Rails
Jing Wang, Yufan Zheng, Yunzhe Ni, Chenren Xu, Feng Qian, Wangyang Li, Wantong Jiang, Yihua Cheng, Zhuo Cheng, Yuanjie Li, Xiufeng Xie, Yi Sun, Zhongfeng Wang
Mobicom 2019
Industry Experience
At Conviva
Research/engineering intern, 2020.10-2021.3, 2022.6-2022.12, 2023.6-now
-
Initial design and implementation of the data processing engine for Time-State Analytics.
-
Content-based video QoE analysis
At MSRA
Research intern, 2021.5-2021.9
- RL-based congestion control algorithm for multi-party video conference applications
At Alibaba
Research intern, 2020.5-2020.9
-
Next generation real-time video platform for DingTalk and TaobaoLive
-
Data-center congestion control algorithm with programmable switches