I am an Assistant Professor of Computer Science at Harvard's John A. Paulson School of Engineering and Applied Sciences, where I am an Associate Faculty at the Kempner Institute and a member of the Theory of Computation group, the ML Foundations group, and the Harvard Quantum Initiative.

I am broadly interested in algorithmic questions about learning from data. In recent years my work has revolved around two threads: developing the mathematical and scientific foundations of diffusion-based generative modeling, and building a theory of learning in quantum systems.

My work has been generously supported by an NSF CAREER award CCF-2441635, an NSF Small (joint with Anurag Anshu) CCF-2430375, an NSF SLES (joint with Boaz Barak and Sham Kakade) IIS-2331831, and the Harvard Dean's Competitive Fund for Promising Scholarship.

Previously I was an NSF postdoc at UC Berkeley under the wise guidance of Prasad Raghavendra. I received my PhD in EECS from MIT as a member of CSAIL and the Theory of Computation group. I was very fortunate to be advised by Ankur Moitra and supported by an MIT Presidential Fellowship and a PD Soros Fellowship. Prior to MIT, I studied mathematics and computer science as an undergraduate at Harvard, where I had the pleasure and honor of working with Salil Vadhan and Leslie Valiant.

Email: sitan (at) seas (dot) harvard (dot) edu
Sitan Chen

Current Group

Former Group Members

Teaching

Recent Papers

Selected Papers

Show All Papers
  1. Parallelism, Critical Windows, and Separations Among Diffusion Language Models [pdf]
    Sitan Chen, Liye Wang
    Manuscript
  2. From Interface to Inference: Eliciting Any-Order Inference from Any-Order Models [pdf]
    Seunggeun Kim, Jaeyeon Kim, Taekyun Lee, Yuyuan Chen, Yilun Du, Sham Kakade, Sitan Chen
    Manuscript
  3. Online Shadow Tomography Matching the Classical Bounds [pdf]
    Sitan Chen, Ryan O'Donnell, Angelos Pelecanos, John Wright
    Manuscript
  4. The Log Log Jam in Gaussian State Tomography [pdf]
    Sitan Chen, Weiyuan Gong, Qi Ye, Zhihan Zhang
    Manuscript
  5. Sublinear Iterations Can Suffice Even for DDPMs [pdf]
    Matthew S. Zhang, Stephen Huan, Jerry Huang, Nicholas M. Boffi, Sitan Chen, Sinho Chewi
    Manuscript
  6. Quantum Probe Tomography [pdf]
    Sitan Chen, Jordan Cotler, Hsin-Yuan Huang
    Manuscript
  7. Selective Underfitting in Diffusion Models [pdf] [website]
    Kiwhan Song, Jaeyeon Kim, Sitan Chen, Yilun Du, Sham Kakade, Vincent Sitzmann
    Manuscript
  8. ReGuidance: A Simple Diffusion Wrapper for Boosting Sample Quality on Hard Inverse Problems [pdf]
    Aayush Karan, Kulin Shah, Sitan Chen
    Manuscript
  9. A Provably Efficient Method for Tensor Ring Decomposition and Its Applications [pdf]
    Han Chen, Sitan Chen, Anru R. Zhang
    SIAM Journal on Mathematics of Data Science
  10. Information-Computation Gaps in Quantum Learning via Low-Degree Likelihood [pdf]
    Sitan Chen, Weiyuan Gong, Jonas Haferkamp, Yihui Quek
    COLT 2026, QIP 2026
  11. Optimal Inference Schedules for Masked Diffusion Models [pdf]
    Sitan Chen, Kevin Cong, Jerry Li
    COLT 2026
  12. Stop Training for the Worst: Progressive Unmasking Accelerates Masked Diffusion Training [pdf]
    Jaeyeon Kim, Jonathan Geuter, David Alvarez-Melis, Sham Kakade, Sitan Chen
    ICML 2026
  13. High-Accuracy and Dimension-Free Sampling with Diffusions [pdf]
    Khashayar Gatmiry, Sitan Chen, Adil Salim
    ICML 2026
    Oral presentation
  14. Fine-Tuning Masked Diffusion for Provable Self-Correction [pdf]
    Jaeyeon Kim, Seunggeun Kim, Taekyun Lee, David Z. Pan, Hyeji Kim, Sham Kakade, Sitan Chen
    ICML 2026
  15. Computation-Utility-Privacy Tradeoffs in Bayesian Estimation [pdf]
    Sitan Chen, Jingqiu Ding, Mahbod Majid, Walter McKelvie
    STOC 2026
  16. Any-Order Flexible Length Masked Diffusion [pdf] [website] [code]
    Jaeyeon Kim, Lee Cheuk-Kit, Carles Domingo-Enrich, Yilun Du, Sham Kakade, Timothy Ngotiaoco, Sitan Chen, Michael Albergo
    ICLR 2026
  17. S4S: Solving for a Diffusion Model Solver [pdf]
    Eric Frankel, Sitan Chen, Jerry Li, Pang Wei Koh, Lillian J. Ratliff, Sewoong Oh
    ICML 2025
  18. Train for the Worst, Plan for the Best: Understanding Token Ordering in Masked Diffusions [pdf]
    Jaeyeon Kim, Kulin Shah, Vasilis Kontonis, Sham Kakade, Sitan Chen
    ICML 2025
    Outstanding Paper Award
  19. Blink of an Eye: A Simple Theory for Feature Localization in Generative Models [pdf]
    Marvin Li, Aayush Karan, Sitan Chen
    ICML 2025
    Oral presentation
  20. Gradient Dynamics for Low-Rank Fine-Tuning Beyond Kernels [pdf] [slides]
    Arif Kerem Dayi, Sitan Chen
    COLT 2025
  21. Predicting Quantum Channels Over General Product Distributions [pdf]
    Sitan Chen, Jaume de Dios Pont, Jun-Ting Hsieh, Hsin-Yuan Huang, Jane Lange, Jerry Li
    COLT 2025
  22. Learning General Gaussian Mixtures with Efficient Score Matching [pdf]
    Sitan Chen, Vasilis Kontonis, Kulin Shah
    COLT 2025
  23. Provably Learning a Multi-Head Attention Layer [pdf] [slides]
    Sitan Chen, Yuanzhi Li
    STOC 2025
  24. Stabilizer Bootstrapping: A Recipe for Agnostic Tomography and Magic Estimation [pdf]
    Sitan Chen, Weiyuan Gong, Qi Ye, Zhihan Zhang
    STOC 2025, QIP 2025
    Short plenary talk
  25. Faster Diffusion-Based Sampling with Randomized Midpoints: Sequential and Parallel [pdf]
    Shivam Gupta, Linda Cai, Sitan Chen
    ICLR 2025
  26. Optimal High-Precision Shadow Estimation [pdf]
    Sitan Chen, Jerry Li, Allen Liu
    QIP 2025, merged with [CLL24a]
  27. Efficient Pauli Channel Estimation with Logarithmic Quantum Memory [pdf] [journal]
    Sitan Chen, Weiyuan Gong
    QIP 2025, PRX Quantum
  28. What Does Guidance Do? A Fine-Grained Analysis in a Simple Setting [pdf]
    Muthu Chidambaram, Khashayar Gatmiry, Sitan Chen, Holden Lee, Jianfeng Lu
    NeurIPS 2024
  29. Unrolled Denoising Networks Provably Learn Optimal Bayesian Inference [pdf]
    Aayush Karan, Kulin Shah, Sitan Chen, Yonina C. Eldar
    NeurIPS 2024
  30. Optimal Tradeoffs for Estimating Pauli Observables [pdf]
    Sitan Chen, Weiyuan Gong, Qi Ye
    FOCS 2024, QIP 2025
    Quanta Magazine, Wired Magazine
  31. A Faster and Simpler Algorithm for Learning Shallow Networks [pdf]
    Sitan Chen, Shyam Narayanan
    COLT 2024
  32. Critical Windows: Non-Asymptotic Theory for Feature Emergence in Diffusion Models [pdf]
    Marvin Li, Sitan Chen
    ICML 2024
  33. An Optimal Tradeoff Between Entanglement and Copy Complexity for State Tomography [pdf]
    Sitan Chen, Jerry Li, Allen Liu
    STOC 2024, QIP 2025
  34. Learning Mixtures of Gaussians Using the DDPM Objective [pdf]
    Kulin Shah, Sitan Chen, Adam R. Klivans
    NeurIPS 2023
  35. The Probability Flow ODE Is Provably Fast [pdf]
    Sitan Chen, Sinho Chewi, Holden Lee, Yuanzhi Li, Jianfeng Lu, Adil Salim
    NeurIPS 2023
  36. When Does Adaptivity Help for Quantum State Learning? [pdf] [slides] [video]
    Sitan Chen, Brice Huang, Jerry Li, Allen Liu, Mark Sellke
    FOCS 2023, QIP 2023
  37. Learning Narrow One-Hidden-Layer ReLU Networks [pdf]
    Sitan Chen, Zehao Dou, Surbhi Goel, Adam R. Klivans, Raghu Meka
    COLT 2023
  38. Restoration-Degradation Beyond Linear Diffusions: A Non-Asymptotic Analysis For DDIM-Type Samplers [pdf]
    Sitan Chen, Giannis Daras, Alexandros G. Dimakis
    ICML 2023
  39. Learning Polynomial Transformations [pdf] [video]
    Sitan Chen, Jerry Li, Yuanzhi Li, Anru R. Zhang
    STOC 2023
  40. Sampling Is as Easy as Learning the Score: Theory for Diffusion Models With Minimal Data Assumptions [pdf] [slides]
    Sitan Chen, Sinho Chewi, Jerry Li, Yuanzhi Li, Adil Salim, Anru R. Zhang
    ICLR 2023
    Oral presentation
  41. Learning to Predict Arbitrary Quantum Processes [pdf] [slides] [journal]
    Hsin-Yuan Huang, Sitan Chen, John Preskill
    QIP 2023, PRX Quantum
  42. The Complexity of NISQ [pdf] [slides] [video] [journal]
    Sitan Chen, Jordan Cotler, Hsin-Yuan Huang, Jerry Li
    QIP 2023, Nature Communications
  43. Learning (Very) Simple Generative Models Is Hard [pdf]
    Sitan Chen, Jerry Li, Yuanzhi Li
    NeurIPS 2022
    Oral presentation
  44. Hardness of Noise-Free Learning for Two-Hidden-Layer Neural Networks [pdf]
    Sitan Chen, Aravind Gollakota, Adam R. Klivans, Raghu Meka
    NeurIPS 2022
    Oral presentation
  45. Tight Bounds for Quantum State Certification with Incoherent Measurements [pdf] [slides] [video]
    Sitan Chen, Brice Huang, Jerry Li, Allen Liu
    FOCS 2022, QIP 2023
  46. Quantum Advantage in Learning From Experiments [pdf] [journal]
    Hsin-Yuan Huang, Michael Broughton, Jordan Cotler, Sitan Chen, Jerry Li, Masoud Mohseni, Hartmut Neven, Ryan Babbush, Richard Kueng, John Preskill, Jarrod R. McClean
    Science
  47. Kalman Filtering with Adversarial Corruptions [pdf]
    Sitan Chen, Frederic Koehler, Ankur Moitra, Morris Yau
    STOC 2022
  48. Minimax Optimality (Probably) Doesn't Imply Distribution Learning for GANs [pdf]
    Sitan Chen, Jerry Li, Yuanzhi Li, Raghu Meka
    ICLR 2022
  49. A Hierarchy for Replica Quantum Advantage [pdf]
    Sitan Chen, Jordan Cotler, Hsin-Yuan Huang, Jerry Li
    QIP 2022, merged with [CCHL21]
  50. Towards Instance-Optimal Quantum State Certification With Independent Measurements [pdf]
    Sitan Chen, Jerry Li, Ryan O'Donnell
    QIP 2022, COLT 2022
    Blurb on Property Testing Review
  51. Symmetric Sparse Boolean Matrix Factorization and Applications [pdf]
    Sitan Chen, Zhao Song, Runzhou Tao, Ruizhe Zhang
    ITCS 2022
  52. Efficiently Learning One Hidden Layer ReLU Networks From Queries [pdf]
    Sitan Chen, Adam R. Klivans, Raghu Meka
    NeurIPS 2021
  53. Exponential Separations Between Learning With and Without Quantum Memory [pdf]
    Sitan Chen, Jordan Cotler, Hsin-Yuan Huang, Jerry Li
    FOCS 2021, QIP 2022
    Invited to SIAM Journal of Computing Special Issue
  54. Online and Distribution-Free Robustness: Regression and Contextual Bandits with Huber Contamination [pdf]
    Sitan Chen, Frederic Koehler, Ankur Moitra, Morris Yau
    FOCS 2021
  55. Learning Deep ReLU Networks Is Fixed-Parameter Tractable [pdf] [video]
    Sitan Chen, Adam R. Klivans, Raghu Meka
    FOCS 2021
  56. Algorithmic Foundations for the Diffraction Limit [pdf] [slides] [code] [video] [Ankur's Simons tutorial]
    Sitan Chen, Ankur Moitra
    STOC 2021
  57. On InstaHide, Phase Retrieval, and Sparse Matrix Factorization [pdf]
    Sitan Chen, Xiaoxiao Li, Zhao Song, Danyang Zhuo
    ICLR 2021
  58. Classification Under Misspecification: Halfspaces, Generalized Linear Models, and Connections to Evolvability [pdf] [code] [Ankur's Simons tutorial]
    Sitan Chen, Frederic Koehler, Ankur Moitra, Morris Yau
    NeurIPS 2020
    Spotlight paper
  59. Learning Structured Distributions from Untrusted Batches: Faster and Simpler [pdf] [code]
    Sitan Chen, Jerry Li, Ankur Moitra
    NeurIPS 2020
  60. Entanglement is Necessary for Optimal Quantum Property Testing [pdf] [slides] [video]
    Sebastien Bubeck, Sitan Chen, Jerry Li
    FOCS 2020
    Blurb on Property Testing Review
  61. Learning Polynomials of Few Relevant Dimensions [pdf] [slides] [video]
    Sitan Chen, Raghu Meka
    COLT 2020
  62. Learning Mixtures of Linear Regressions in Subexponential Time via Fourier Moments [pdf] [slides] [video]
    Sitan Chen, Jerry Li, Zhao Song
    STOC 2020
  63. Efficiently Learning Structured Distributions from Untrusted Batches [pdf] [slides] [video]
    Sitan Chen, Jerry Li, Ankur Moitra
    STOC 2020
  64. Improved Bounds for Sampling Colorings via Linear Programming [pdf] [slides]
    Sitan Chen, Michelle Delcourt, Ankur Moitra, Guillem Perarnau, Luke Postle
    (merger of [CM18] and [DPP18])
    SODA 2019
  65. Beyond the Low-Degree Algorithm: Mixtures of Subcubes and Their Applications [pdf] [slides]
    Sitan Chen, Ankur Moitra
    STOC 2019
  66. Basis Collapse For Holographic Algorithms over All Domain Sizes [pdf] [slides] [video]
    Sitan Chen
    STOC 2016
  67. Pseudorandomness for Read-Once, Constant-Depth Circuits [pdf]
    Sitan Chen, Thomas Steinke, Salil Vadhan
    Manuscript

Thesis

Service

Other