Scholar
Tianyu Pang
Google Scholar ID: xcDPmRMAAAAJ
Dartmouth College
Citations & Impact
Citations
35
H-index
3
i10-index
1
Publications
8
Co-authors
11
list available
Publications
Resume (English only)
Academic Achievements
- Publications: 'From Spikes to Heavy Tails: Unveiling the Spectral Evolution of Neural Networks' accepted by TMLR; 'LIFT the Veil for the Truth: Principal Weights Emerge after Rank Reduction for Reasoning-Focused Supervised Fine-Tuning' accepted by ICML 2025 as Poster; 'Model Balancing Helps Low-data Training and Fine-tuning' accepted by EMNLP 2024 as Oral Presentation; 'Temperature Balancing, Layer-wise Weight Analysis, and Neural Network Training' accepted by NeurIPS 2023 as Spotlight.
Research Experience
- Completed a wonderful three-month visiting at Dartmouth College in October 2024.
Education
- 2025.09 - present, Ph.D. in Machine Learning, Department of Computer Science, Dartmouth College, Advisor: Prof. Yaoqing Yang; 2022.09 - 2025.06, M.S. in Mathematics, Department of Mathematics, Nanjing University; 2018.09 - 2022.06, B.S. in Statistics, Kuang Yaming Honors School, Nanjing University.
Background
- Research Interests: Model diagnostics and mechanistic interpretability. Current research focuses on understanding the mechanisms, dynamics, and generalization of LLMs from the perspective of random matrix theory, high-dimensional statistics, and loss landscape; leveraging model/data diagnostics and interpretability insights to improve the transparency, robustness, and efficiency of (scientific) machine learning.
