Nate S. Woodward
ML + Theoretical Physics
I am currently a Research Fellow / Member of Technical Staff Intern at First Principles, on leave from my PhD at the University of Wisconsin–Madison until Spring 2027.
My PhD research at UW–Madison focuses on machine learning techniques for theoretical physics, particularly improving LLM reasoning for theoretical physics.
I graduated from MIT in 2025 with degrees in Physics and Mathematics. During my undergraduate studies I worked in Professor Phil Harris's group and with the NSF AI Institute for Artificial Intelligence and Fundamental Interactions (IAIFI), building ML tools for high‑energy physics. My experience spans geometric ML for jet physics, detector readout algorithms, and theoretical QFT calculations.
Recent Publications
• Fine-Tuning Small Reasoning Models for Quantum Field Theory (2026)
• AutoSciDACT: Automated Scientific Discovery through Contrastive Embedding and Hypothesis Testing (2025)
• Re-Simulation-based Self-Supervised Learning for Pre-Training Foundation Models (2025)
Recent Posts
• Tailoring Theory to Experiment: An Interview with Prof. Ian Moult (2025-02-17)
• Cancer @ 20 (2023-01-18)
Recent Talks
• LITP Spring Symposium: Fine-Tuning Small Reasoning Models for Quantum Field Theory (2026)
• Bites of Foundation Models for Science: Product Manifold Machine Learning for Physics (2024)