Synthesizing Physics Formulae with Transformers

arXiv:2610.03947 · cs.LG, cs.SC · Submitted 2026-10-02 · Read on arXiv

cs.LG, cs.SC

Submitted: 2026-10-02

Updated: 2026-10-02

Code: https://github.com/ShwayW/shorter-formulae

Terminology

• Transformer — 1× in this paper · explained in 2 episodes

Sources

• Analyzing Generalization in Pre-Trained Symbolic Regression
• Interpretable Machine Learning for Science with PySR and SymbolicRegression.jl
• Finetuning Large Language Model as an Effective Symbolic Regressor
• Formal Algorithms for Transformers
• Gaussian Error Linear Units (GELUs)
• A Transformer Model for Symbolic Regression towards Scientific Discovery
• The Llama 3 Herd of Models
• Qwen2.5-Coder Technical Report

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks