Yanghui Wu | Computer Science PhD Candidate

I am a graduate student in Computer Science at the University of Rochester, advised by Professor Chen Ding. My research focuses on program locality, cache systems, and performance analysis. I am particularly interested in analytical and compiler-based techniques for understanding memory behavior, including symbolic locality analysis, cache modeling, and cache-management policies.

Recent Publications

SALT: Symbolic Analysis of Loop Tiling

Yanghui Wu, Yifan Zhu, Yekai Pan, and Chen Ding.

International Conference on Parallel Architectures and Compilation Techniques (PACT 2026). Best Paper Candidate (Plenary Session).
PDF HTML PACT 2026 Technical Program Artifact DOI: 10.1145/3838684.3846873
Benchmarking Cache Programming Against Optimal Caching

Yanghui Wu, Vincent Michelini, Chen Ding, and Dorin Patru.

International Symposium on Memory Systems (MemSys 2025), pages 271–281.
DOI
Deriving Symbolic AET Using Locality Identity Equation

Fangzhou Liu, Yifan Zhu, Chen Ding, Yekai Pan, and Yanghui Wu.

IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW 2026), pages 300–309.
DOI
Fully Symbolic Analysis of Loop Locality: Using Imaginary Reuse to Infer Real Performance

Yifan Zhu, Yekai Pan, Chen Ding, and Yanghui Wu.

arXiv preprint arXiv:2603.10196, 2026.
arXiv

Presentations & Talks

Symbolic Analysis of Loop Tiling for Memory Locality and Cache Optimization

Upstate Programming Languages Seminar (Upstate PL), Buffalo, NY, August 15, 2026.

Research Projects

Symbolic Locality and Cache Analysis for Rectangular Loop Programs

Symbolic analysis of locality and cache behavior for rectangular loop programs, including tiled and transformed loops, using compiler and program-analysis techniques.

Program-Guided Cache Management with Per-Reference Leases

Program-guided cache-management techniques that use per-reference leases to make cache replacement decisions.

Teaching Assistant

CSC 254/454: Programming Language Design & Implementation

Fall 2022, Fall 2023, Fall 2024 — Instructor: Michael L. Scott

Spring 2024 — Instructor: George Ferguson

CSC 252/452: Computer Organization

Fall 2021 — Instructor: H. Alan Beadle

Departmental Service

Creator of the Computer Science Department 50th Anniversary Timeline, featured during the 2024 Meliora weekend and published in the department's annual Multicast newsletter.