| CARVIEW |
Select Language
HTTP/2 200
server: GitHub.com
content-type: text/html; charset=utf-8
last-modified: Tue, 11 Nov 2025 01:41:43 GMT
access-control-allow-origin: *
strict-transport-security: max-age=31556952
etag: W/"69129457-2e87"
expires: Mon, 29 Dec 2025 00:27:44 GMT
cache-control: max-age=600
content-encoding: gzip
x-proxy-cache: MISS
x-github-request-id: 3279:2BC55:819B1F:919269:6951C8A7
accept-ranges: bytes
age: 0
date: Mon, 29 Dec 2025 00:17:44 GMT
via: 1.1 varnish
x-served-by: cache-bom-vanm7210027-BOM
x-cache: MISS
x-cache-hits: 0
x-timer: S1766967464.345726,VS0,VE202
vary: Accept-Encoding
x-fastly-request-id: 9bed5fe607e508f03d7adf3babbfe14bec46eceb
content-length: 4148
Mengdi Wu
About Me
I am Mengdi Wu (吴梦迪), a third-year Ph.D. student in the Computer Science Department at Carnegie Mellon University, advised by Zhihao Jia. I received my undergraduate degree from the Yao Class at Tsinghua University in 2023. I am honored to be supported by the Amazon Fellowship for my research.
I work on machine learning systems, with a focus on ML compilers and superoptimization. I am interested in understanding the computational structure of machine learning applications and developing techniques to make them more efficient and reliable.
Research Experience
| 2022.02~2022.08 | I visited CMU as a research intern, where I worked with Zhihao Jia on accelerating distributed DL training. |
|---|---|
| 2020.11~2021.11 | I joined ARChip Lab led by Kaisheng Ma at Tsinghua University, and worked on designing architectures and schedulers for DNN accelerators. |
Awards
-
National Student Computer System Capability Challenge 2020 (Compiler track), 2nd Price [Cutepiler]
With Jiatu Li, Runda Liu, and Zhidong Wang.
- National Olympiad in Informatics, China, Gold Medal
Publications
- Identity Testing for Circuits with Exponentiation Gates
- Jiatu Li*, Mengdi Wu*
- ITCS 2026
- Collie: Token-Level Co-Serving of LLM Inference and Fine-Tuning with SLO Guarantees
- Gabriele Oliaro, Xupeng Miao, Xinhao Cheng, Vineeth Kada, Ruohan Gao, Yingyi Huang, Remi Delacourt, April Yang, Yingcheng Wang, Mengdi Wu, Colin Unger, Zhihao Jia
- NSDI 2026
- Mirage: A Multi-Level Superoptimizer for Tensor Programs
- Mengdi Wu, Xinhao Cheng, Shengyu Liu, Chunan Shi, Jianan Ji, Kit Man Ao, Praveen Velliengiri, Xupeng Miao, Oded Padon, Zhihao Jia
- OSDI 2025
- GraphPipe: Improving the Performance and Scalability of DNN Training with Graph Pipeline Parallelism
- Byungsoo Jeon*, Mengdi Wu*, Shiyi Cao*, Sunghyun Kim*, Sunghyun Park, Neeraj Aggarwal, Colin Unger, Daiyaan Arfeen, Peiyuan Liao, Xupeng Miao, Mohammad Alizadeh, Gregory R. Ganger, Tianqi Chen, Zhihao Jia
- ASPLOS 2025
- LACO: A Latency-Constraint Offline Neural Network Scheduler towards Reliable Self-Driving Perception
- Zhanhong Tan, Zijian Zhu, Mengdi Wu, Kaisheng Ma
- ICCAD 2024
- Finding the Task-Optimal Low-Bit Sub-Distribution in Deep Neural Networks
- Runpei Dong*, Zhanhong Tan*, Mengdi Wu, Linfeng Zhang, Kaisheng Ma
- ICML 2022