Sunny Qin

Tian (Sunny) Qin

PhD Student, Harvard CS

I am a third-year PhD student at Harvard, advised by Sham Kakade and David Alvarez-Melis, and a member of the Machine Learning Foundations Group. I'm honored to have been selected as part of the 2025 Apple Scholars in AI/ML. My research centers around data-centric AI and the science underlying foundational models. Specifically, I focus on developing synthetic data generation techniques to enhance model capabilities and utilizing sandbox approaches to better understand these models' behaviors.

Blog Posts

To Backtrack or Not to Backtrack: When Sequential Search Limits Model Reasoning

Distinguishing the Knowable from the Unknowable with Language Models

Research

To Backtrack or Not to Backtrack: When Sequential Search Limits Model Reasoning. arXiv preprint

Tian Qin, David Alvarez-Melis, Samy Jelassi, Eran Malach.

[ArXiv]

Distributional Scaling Laws for Emergent Capabilities. arXiv preprint

Rosie Zhao, Tian Qin, David Alvarez-Melis, Sham Kakade, Naomi Saphra.

[ArXiv]

Sometimes I am a Tree: Data Drives Fragile Hierarchical Generalization. NeurIPS 2024 Workshop on Scientific Methods for Understanding Deep Learning

Tian Qin, Naomi Saphra, David Alvarez-Melis.

Distinguishing the Knowable from the Unknowable with Language Models. ICML 2024

Tian Qin, Gustaf Ahdritz, Nikhil Vyas, Boaz Barak, Benjamin L. Edelman.

[ArXiv] [GitHub]

A Label is Worth A Thousand Images in Dataset Distillation. NeurIPS 2024

Tian Qin, Zhiwei Deng, David Alvarez-Melis.

[ArXiv] [GitHub]

Distributional Dataset Distillation with Subtask Decomposition. ICLR 2024 Workshops

Tian Qin, Zhiwei Deng, David Alvarez-Melis.

[ArXiv] [GitHub]

Meta-PDE: Learning to Solve PDEs Quickly Without a Mesh. Preprint

Tian Qin, Alex Beatson, Deniz Oktay, Nick McGreivy, Ryan P. Adams.

[ArXiv] [GitHub]

Teaching & Service

I am a mentor for the UWC Science Collective, where I volunteer to mentor passionate high school students to do research.
Invited talk: "Data Drives Unstable Hierarchical Generalization in LMs" . Seminars on Formal Languages and Neural Networks March 2025.
Invited talk: "A Label is Worth a Thousand Images in Dataset Distillation". Voxel 51 Boston AI, ML and Computer Vision Meetup Feb 2025.
CS181: Machine Learning, Fall 2023 (TF)

Personal

Outside research, I enjoy spending my free time outdoors — climbing, and just recently started mountaineering! I am a board member and an education officer of the Harvard Mountaineering Club.

Contact

Download my CV (PDF)

Email: tqin[AT]g.harvard.edu

Feel free to reach out - I am always open for collaborations, discussing research ideas!