Ranran Shen

Ranran Shen

Ph.D. student in Computer Science

Hi there! I am Ranran Shen (申冉冉), a Ph.D. student in Computer Science at University of Science and Technology of China (USTC). I feel incredibly fortunate to be advised by Prof. Pan Peng, whose guidance has been instrumental in shaping my academic journey. I am a member of the AntBag group. Previously, I got my Bachelor’s degree in Computer Science and Technology from Central South University (CSU) in 2022, where I was advised by Prof. Ying Zhao.

My research interests lie broadly in sublinear-time algorithms and graph algorithms. Specifically, I am currently focusing on sublinear-time clustering algorithms on graphs, though I am still in the process of forming a more defined understanding of my research direction.

If you’d like to discuss my research or have any questions, please feel free to email me at: ranranshen@mail.ustc.edu.cn. I’m also happy to connect with people both within and beyond academia🔅.

🔥 News

  • 2026.01:  🎉🎉 One paper is accepted by ICLR’2026 (main conference) and many thanks to my co-authors!📝
  • 2024.04:  🎉🎉 Thrilled to start an internship at Tencent, focusing on GUI agent.💻
  • 2023.12:  🎉🎉 Incredibly excited to attend the NeurIPS’2023 conference (hosted in New Orleans, Louisiana, USA) and present my poster.✈️
  • 2023.10:  🎉🎉 I’m honored to receive the National Scholarship!🏆
  • 2023.09:  🎉🎉 One paper is accepted by NeurIPS’2023 (main conference)!📝

📝 Publications

ICLR 2026
sym

Sublinear Spectral Clustering Oracle with Little Memory (ICLR’2026)

Ranran Shen, Xiaoyi Zhu, Pan Peng, Zengfeng Huang

Paper and Poster will be updated soon.

  • We study the problem of designing sublinear spectral clustering oracles for well-clusterable graphs. A major limitation of existing oracles is that constructing $\mathcal{D}$ requires $\Omega(\sqrt{n})$ memory, which becomes a bottleneck for massive graphs and memory-limited settings. In this paper, we break this barrier and establish a memory-time trade-off for sublinear spectral clustering oracles.
NeurIPS 2023
sym

A Sublinear-Time Spectral Clustering Oracle with Improved Preprocessing Time (NeurIPS’2023)

Ranran Shen, Pan Peng

Paper｜Slides: short (5 min), full｜Poster

  • We address the problem of designing a sublinear-time spectral clustering oracle for graphs that exhibit strong clusterability. Previous oracles have relied on either a $\textrm{poly}(k)\cdot\log n$ gap between inner and outer conductances or exponential (in $k/\varepsilon$) preprocessing time. Our algorithm relaxes these assumptions, albeit at the cost of a slightly higher misclassification ratio.

📖 Education

💻 Internships

  • 2024.04 ~ 2024.08, Tencent , Shenzhen.

🏆 Honors and Awards

  • 2026.03, ICLR 2026 Travel Grant.
  • 2023.10, National Scholarship (Top 2%)🌷.
  • 2022 ~ 2025, First Prize, USTC Graduate Student Academic Scholarship.
  • 2022.06, Honored as The Outstanding Bachelor Graduate of Hunan Province (Top 3%).
  • 2022.06, Honored as The Outstanding Bachelor Graduate of CSU (Top 15%).
  • 2021.10, National Scholarship (Undergraduate, Top 2%)🌷.
  • 2019 ~ 2021, Honored as The Outstanding Student of School of Computer Science and Engineering, CSU.
  • 2020 ~ 2021, First Prize, CSU Undergraduate Student Academic Scholarship.
  • 2019.10, Second Prize, CSU Undergraduate Student Academic Scholarship.

🎖 Services

  • Invited Reviewer:
    • ICLR 2026, ICML 2026.
  • Teaching Assistant:

🦕 Special Links

  • Here is the link to the homepage of my boyfriend: Zhaoyi Li.