Antonin Sulc

Antonin Sulc, Ph.D.

AI for Autonomous Particle Accelerators

Lawrence Berkeley National Laboratory

Download CV

About

I am a Researcher at Lawrence Berkeley National Laboratory (LBNL) developing intelligent algorithms and autonomous control systems for large-scale scientific infrastructure. My work bridges large language models, agentic systems, and high-energy physics to enable self-optimizing operations in safety-critical environments.

Previously a Senior Data Scientist at the Helmholtz Association (DESY/XFEL), I pioneered the use of domain-specific language models in scientific contexts. Holding a Ph.D. in Computer Vision from the University of Konstanz, I actively apply the mathematical rigor of high-dimensional signal processing to extract actionable insights from the complex, multi-modal machine states of modern particle accelerators.

Software & Open Source

event2vector

A scikit-learn style Python library providing a geometric approach to learning composable, highly interpretable representations of discrete event sequences.

torchmodal

A framework introducing Differentiable Modal Logic for PyTorch—enabling the training of Modal Logical Neural Networks (MLNNs).

Osprey

A scalable, production-ready framework for orchestrating agentic AI in safety-critical facility operations. Deployed successfully at the Advanced Light Source.

PACuna (LLM_NeuralIPS23)

Automated data collection and fine-tuning pipelines designed to adapt Large Language Models securely to the complex domains of particle accelerators.

Publications

Solving PDEs in One Shot via Fourier Features with Exact Analytical Derivatives

A. Sulc

arXiv preprint, 2026

Paper

Differentiable Logical Programming for Quantum Circuit Discovery and Optimization

A. Sulc

arXiv preprint, 2026

Paper

Modal Logical Neural Networks

A. Sulc

arXiv preprint, 2025

Paper

Agentic artificial intelligence for multistage physics experiments at a large-scale user facility particle accelerator

T. Hellert, D. Bertwistle, S. C. Leemann, A. Sulc, M. Venturini

Physical Review Research, 2026

Paper

Osprey: Production-ready agentic AI for safety-critical control systems

T. Hellert, J. Montenegro, A. Sulc

APL Machine Learning, 2026

Paper

Quantum Noise Tomography with Physics-Informed Neural Networks

A. Sulc

NeurIPS ML4Physics Workshop, 2025

Paper

eLog Analysis for Accelerators: Status and Future Outlook

A. Sulc, T. Hellert, A. Reed, A. Carpenter et al.

16th International Particle Accelerator Conference (IPAC'25)

Paper

QCD in Language Models: What They Really Know About QCD?

A. Sulc, P. Connor

EPS-High Energy Physics, 2025

Paper

ChatQCD: Let Large Language Models Explore QCD

A. Sulc, P. Connor

42nd International Conference on High Energy Physics (ICHEP 2024)

Towards Agentic AI on Particle Accelerators

A. Sulc, T. Hellert, R. Kammering, H. Hoschouer, J. St. John

NeurIPS ML4Physics Workshop, 2024

Paper

Towards Unlocking Insights from Logbooks Using AI

A. Sulc, G. Hartmann, J. Maldonado, V. Kain et al.

15th International Particle Accelerator Conference (IPAC'24)

Paper

PACuna: Automated Fine-Tuning of Language Models for Particle Accelerators

A. Sulc, R. Kammering, A. Eichler, T. Wilksen

NeurIPS ML4Physics Workshop, 2023

Paper

Unsupervised Log Anomaly Detection with Few Unique Tokens

A. Sulc, A. Eichler, T. Wilksen

IET Information Security, 2023

Paper

Depth from Spectral Defocus Blur

S. Ishihara, A. Sulc, I. Sato

IEEE International Conference on Image Processing (ICIP), 2019

Paper

Depth Estimation Using Spectrally Varying Defocus Blur

S. Ishihara, A. Sulc, I. Sato

Journal of the Optical Society of America A, 2021

Paper

What Sparse Light Field Coding Reveals About Scene Structure

O. Johannsen, A. Sulc, B. Goldluecke

IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016

Paper

Contact