Ali Shehper

I am an AI researcher, currently a Senior Research Scientist on the Nemotron team at NVIDIA.

I was trained as a physicist, studying supersymmetric and topological quantum field theories. I switched to studying Reinforcement Learning for Math during my postdoc studies, and most recently, I have worked on language models at Essential AI.

Blog

• Revisiting Neural Network Parameterizations for Optimal Performance
• A Features’ Perspective on Neural Scaling Laws

Papers

Science of Deep Learning

• A Training-Time Diagnostic for Generalization via the Log-Alignment Ratio.
  A. Shehper, A. Vaswani.

Physics

• Nonabelian Twists of the D4 Theory
  J. Distler, B. Ergun, A. Shehper (lead contribution with B. Ergun, alphabetical order).

• Symmetries of 2d TQFTs and Equivariant Verlinde Formulae for General Groups
  S. Gukov, D. Pei, C. Reid, A. Shehper (lead contributor, alphabetical order).

• Distinguishing 4d N=2 SCFTs
  J. Distler, B. Ergun, A. Shehper (equal contribution with B. Ergun, alphabetical order).

• Deformations of surface defect moduli spaces
  A. Neitzke, A. Shehper (lead contributor, alphabetical order).

• PhD Thesis: Aspects of Supersymmetric and Topological Quantum Field Theories.

RL For Math

• What makes math problems hard for reinforcement learning: a case study
  (Accepted at NeurIPS 2025)
  A. Shehper, A. M. Medina-Mardones, L. Fagan, B. Lewandowski, A. Gruen, Y. Qiu, P. Kucharski, Z. Wang, S. Gukov.

• The Two-Hump Problem: Bridging the Difficulty Gap in Mathematical Reinforcement Learning
  (Accepted at ICML 2026)
  L. Fagan*, M. Tarquini*, A. Shehper*, M. Manko, A. Gruen, C. Huang, G. Butbaia, D. Passaro, S. Gukov. (* equal contribution)

• Hierarchical Reinforcement Learning for Sparse-Reward Search in Commutative Algebra
  (Accepted at ICML 2026)
  G. Butbaia, P. Orland, C. Huang, D. Passaro, L. Fagan, M. Tarquini, H. Dao, D. Eisenbud, A. Shehper, S. Gukov.

Code

• AC-Solver A Python library for tackling long-horizon, ultra-sparse-reward RL environments, designed to accompany our case study.

• Sparse-Dictionary-Learning An open-source implementation of Anthropic’s Towards Monosemanticity: Decomposing Language Models with Dictionary Learning.

• Neural Scaling Laws An implementation of Scaling Laws for Neural Language Models, along with results from An Empirical Model of Large-Batch Training.

• Language Model Feature Browser A visualizer for features learned by a 1-layer Language Model, with the GitHub Repository.

Google Scholar / GitHub / Twitter / LinkedIn