Scholar
Mingjie Sun
Google Scholar ID: wCZbouUAAAAJ
Thinking Machines Lab
Citations & Impact
Citations
3,639
H-index
13
i10-index
14
Publications
20
Co-authors
7
list available
Contact
Publications
Resume (English only)
Academic Achievements
- Published multiple papers including 'Idiosyncrasies in Large Language Models' (ICML 2025), 'Test-Time Adaptation Induces Stronger Accuracy and Agreement-on-the-Line' (NeurIPS 2024), 'Massive Activations in Large Language Models' (COLM 2024, Oral at ICLR 2024 ME-FoMo Workshop), 'A Simple and Effective Pruning Approach for Large Language Models' (ICLR 2024), 'Single Image Backdoor Inversion via Robust Smoothed Classifiers' (CVPR 2023), '(Certified!!) Adversarial Robustness for Free!' (ICLR 2023), 'Test-Time Adaptation via Conjugate Pseudo-Labels' (NeurIPS 2022), 'Denoised Smoothing: A Provable Defense for Pretrained Classifiers' (NeurIPS 2020), 'Rethinking the Value of Network Pruning' (ICLR 2019, Best Paper Award at NeurIPS 2018 Compact Deep Neural Networks Workshop).
Research Experience
- Works at Thinking Machines Lab, involved in multiple research projects such as idiosyncrasies in large language models, test-time adaptation, and massive activations.
Education
- Ph.D. in Computer Science from Carnegie Mellon University, advised by Zico Kolter; B.S. in Computer Science from Yao Class, Tsinghua University.
Background
- Currently a Member of Technical Staff at Thinking Machines Lab. Research interests include idiosyncrasies in large language models, test-time adaptation, and massive activations.
Miscellany
- Chinese name: 孙明杰
