Tian Zhang (张添)

tianzh [at] seas.upenn.edu  ·  CV

I'm a fifth-year PhD student in Computer and Information Science at the University of Pennsylvania. I am fortunately advised by Erik Waingarten. I received my bachelor's in computer science and statistics from the University of Michigan - Ann Arbor, where Thatchaphol Saranurak was my undergraduate research advisor.

I am interested in algorithms for massive datasets, such as similarity search. I am also interested in machine learning more broadly, including LLM/VLM post-training, which I worked on as an applied scientist intern at Amazon.

I'm on the job market. I graduate in January 2027 and am looking for research scientist and applied scientist roles in retrieval, vector search, and LLM/VLM post-training. My CV is here.

News

• Sep 2026Two new preprints on arXiv: fast geometric spanners via ANN search, and approximate nearest neighbor in ultra-high dimensional ℓ_∞.
• Sep 2026Prune, Don't Rebuild accepted to NeurIPS 2026.
• Summer 2026Applied Scientist Intern at Amazon in Seattle.
• Feb 2026Directed Euclidean spanners paper accepted to STOC 2026.
• Spring 2026Helped develop CIS 5030: Algorithms for Big Data at Penn.

Publications

• Learn Where to Store: Query-Aware IVF
  Zack Ives, Jiaming Liang, Ashwin Padaki, Erik Waingarten, Tian Zhang
  In submission
• Fast Geometric Spanners via Approximate Nearest Neighbor Search
  Alexandr Andoni, Manuel Paez, Krish Singal, Tian Zhang
  In submission [arXiv]
• Prune, Don't Rebuild: Efficiently Tuning α-Reachable Graphs for Nearest Neighbor Search
  Tian Zhang, Ashwin Padaki, Jiaming Liang, Zack Ives, Erik Waingarten
  NeurIPS 2026 [arXiv]
• Approximate Nearest Neighbor in Ultra-High Dimensional ℓ_∞
  Nathan White, Tian Zhang
  In submission [arXiv]
• Near-Optimal Directed Euclidean Spanners in High Dimensions
  Rajesh Jayaram, Shyamal Patel, Cliff Stein, Erik Waingarten, Tian Zhang
  STOC 2026 [DOI]
• Average-Distortion Sketching
  Yiqiao Bao, Anubhav Baweja, Nicolas Menand, Erik Waingarten, Nathan White, Tian Zhang
  FOCS 2025 [arXiv]
• Data-Dependent LSH for the Earth Mover's Distance
  Rajesh Jayaram, Erik Waingarten, Tian Zhang
  STOC 2024 [arXiv]
• Vertex Sparsifiers for Hyperedge Connectivity
  Han Jiang, Shang-En Huang, Thatchaphol Saranurak, Tian Zhang
  ESA 2022 [arXiv]

Feel free to email me for manuscripts of papers that have not been posted yet.

Industry Experience

• Amazon, Applied Scientist Intern, IP Protection (Seattle, May–Aug 2026)
  Post-trained vision-language models for IP infringement detection.

Teaching

• Course development, CIS 5030: Algorithms for Big Data, Penn, Spring 2026. Instructors: Anindya De and Erik Waingarten.
• Teaching assistant, CIS 6770: Algorithms for Big Data, Penn, Spring 2025. Instructor: Erik Waingarten.
• Teaching assistant, CIS 3200: Introduction to Algorithms, Penn, Fall 2023. Instructor: Erik Waingarten.
• Grader, EECS 477: Introduction to Algorithms, University of Michigan, Fall 2021. Instructor: Greg Bodwin.