- Physical Review Journals
All JournalsPhysics Magazine

Physical Review X

Reuse & Permissions

It is not necessary to obtain permission to reuse this article or its components as it is available under the terms of the Creative Commons Attribution 4.0 International license. This license permits unrestricted use, distribution, and reproduction in any medium, provided attribution to the author(s) and the published article's title, journal citation, and DOI are maintained. Please note that some figures may have been included with permission from other third parties. It is your responsibility to obtain the proper permission from the rights holder directly for these figures.

Export citation

Export citation

Choose format for download:

Download Citation
  • Open Access

Agentic Exploration of Physics Models

Maximilian Nägele* and Florian Marquardt

  • *Contact author: maximilian.naegele@mpl.mpg.de
PDF

Phys. Rev. X 16, 031002 – Published 8 July, 2026

DOI: https://doi.org/10.1103/xnqc-q6nt

Abstract

The process of scientific discovery relies on an interplay of observations, analysis, and hypothesis generation. Machine learning is increasingly being adopted to address individual aspects of this process. However, it remains an open challenge to fully automate the heuristic, iterative loop required to discover the laws of an unknown system by exploring it through experiments and analysis, without tailoring the approach to the specifics of a given task. Here, we introduce sciexplorer, an agent that leverages large language model tool-use capabilities to enable exploration of systems without any domain-specific blueprints and apply it to physical systems that are initially unknown to the agent. We test sciexplorer on a broad set of models spanning mechanical dynamical systems, wave evolution, and quantum many-body physics. Despite using a minimal set of tools, primarily based on code execution, we observe impressive performance on tasks such as recovering equations of motion from observed dynamics and inferring Hamiltonians from expectation values. The demonstrated effectiveness of this setup opens the door toward similar scientific exploration in other domains, without the need for fine-tuning or task-specific instructions.

View figure in article

Physics Subject Headings (PhySH)

Popular Summary

At the core of the natural sciences is the iterative exploration of unknown systems through experiments, analysis, and hypothesis generation. Here, we present sciexplorer, an agentic artificial scientist built on a large language model that, for the first time, automates this process without tailoring to specific tasks or systems. sciexplorer operates with a minimal set of generic instructions and code-based general-purpose tools to autonomously generate a heuristic multistep workflow. Thanks to its generalist approach, sciexplorer performs impressively across diverse physical domains, including classical mechanics, waves, and quantum many-body physics. For instance, it successfully recovers equations of motion from observed dynamics and infers Hamiltonians from expectation values. In the future, sciexplorer could be applied to modern physics experiments that are controlled through code-based interfaces.

Article Text

Supplemental Material

References (56)

  1. Alexey A. Melnikov, Hendrik Poulsen Nautrup, Mario Krenn, Vedran Dunjko, Markus Tiersch, Anton Zeilinger, and Hans J. Briegel, Active learning machine learns to create new quantum experiments, Proc. Natl. Acad. Sci. U.S.A. 115, 1221 (2018).
  2. Nathan J. Szymanski, Bernardus Rendy, Yuxing Fei, Rishi E. Kumar, Tanjin He, David Milsted, Matthew J. McDermott, Max Gallant, Ekin Dogus Cubuk, Amil Merchant, Kim et al., An autonomous laboratory for the accelerated synthesis of novel materials, Nature (London) 624, 86 (2023).
  3. Xiaobo Li, Yu Che, Linjiang Chen, Tao Liu, Kewei Wang, Lunjie Liu, Haofan Yang, Edward O. Pyzer-Knapp, and Andrew I. Cooper, Sequential closed-loop Bayesian optimization as a guide for organic molecular metallophotocatalyst formulation discovery, Nat. Chem. 16, 1286 (2024).
  4. John Jumper, Richard Evans, Alexander Pritzel, Tim Green, Michael Figurnov, Olaf Ronneberger, Kathryn Tunyasuvunakool, Russ Bates, Augustin Žídek, Anna Potapenko et al., Highly accurate protein structure prediction with AlphaFold, Nature (London) 596, 583 (2021).
  5. Amil Merchant, Simon Batzner, Samuel S. Schoenholz, Muratahan Aykol, Gowoon Cheon, and Ekin Dogus Cubuk, Scaling deep learning for materials discovery, Nature (London) 624, 80 (2023).
  6. Annabelle Bohrdt, Christie S. Chiu, Geoffrey Ji, Muqing Xu, Daniel Greif, Markus Greiner, Eugene Demler, Fabian Grusdt, and Michael Knap, Classifying snapshots of the doped Hubbard model with machine learning, Nat. Phys. 15, 921 (2019).
  7. Hugo Dalla-Torre, Liam Gonzalez, Javier Mendoza-Revilla, Nicolas Lopez Carranza, Adam Henryk Grzywaczewski, Francesco Oteri, Christian Dallago, Evan Trop, Bernardo P. de Almeida, Hassan Sirelkhatim et al., Nucleotide transformer: Building and evaluating robust foundation models for human genomics, Nat. Methods 22, 287 (2025).
  8. Till Richter, Mojtaba Bahrami, Yufan Xia, David S. Fischer, and Fabian J. Theis, Delineating the effective use of self-supervised learning in single-cell genomics, Nat. Mach. Intell. 7, 68 (2025).
  9. Miles Cranmer, Alvaro Sanchez-Gonzalez, Peter Battaglia, Rui Xu, Kyle Cranmer, David Spergel, and Shirley Ho, Discovering symbolic models from deep learning with inductive biases, in Proceedings of the 34th International Conference on Neural Information Processing Systems, NIPS ’20 (Curran Associates Inc, Red Hook, NY, 2020), ISBN [Amazon][WorldCat], https://proceedings.neurips.cc/paper_files/paper/2020/file/c9f2f917078bd2db12f23c3b413d9cba-Paper.pdf.
  10. Takeshi Kojima, Shixiang Shane Gu, Machel Reid, Yutaka Matsuo, and Yusuke Iwasawa, Large language models are zero-shot reasoners, Advances in Neural Information Processing Systems (2022), Vol. 35, pp. 22199–22213, https://proceedings.neurips.cc/paper_files/paper/2022/hash/8bb0d291acd4acf06ef112099c16f326-Abstract-Conference.html.
  11. Jean-Baptiste Alayrac, Jeff Donahue, Pauline Luc, Antoine Miech, Iain Barr, Yana Hasson, Karel Lenc, Arthur Mensch, Katie Millicah, Malcolm Reynolds et al., Flamingo: A visual language model for few-shot learning, in Proceedings of the 36th International Conference on Neural Information Processing Systems, NIPS ’22 (Curran Associates Inc, Red Hook, NY, 2022), ISBN [Amazon][WorldCat].
  12. Bernardino Romera-Paredes, Mohammadamin Barekatain, Alexander Novikov, Matej Balog, M. Pawan Kumar, Emilien Dupont, Francisco J. R. Ruiz, Jordan S. Ellenberg, Pengming Wang, Omar Fawzi et al., Mathematical discoveries from program search with large language models, Nature (London) 625, 468 (2024).
  13. Mark Chen, Jerry Tworek, Heewoo Jun, Qiming Yuan, Henrique Ponde de Oliveira Pinto, Jared Kaplan, Harri Edwards, Yuri Burda, Nicholas Joseph, Greg Brockman et al., Evaluating large language models trained on code, arXiv:2107.03374.
  14. Lei Huang, Weijiang Yu, Weitao Ma, Weihong Zhong, Zhangyin Feng, Haotian Wang, Qianglong Chen, Weihua Peng, Xiaocheng Feng, Bing Qin, and Ting Liu, A survey on hallucination in large language models: Principles, taxonomy, challenges, and open questions, ACM Trans. Inf. Syst. 43, 42 (2025).
  15. Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik Narasimhan, and Yuan Cao, React: Synergizing reasoning and acting in language models, in Proceeding sof the International Conference on Learning Representations (2023), https://openreview.net/forum?id=WE_vluYUL-X.
  16. John Yang, Carlos E. Jimenez, Alexander Wettig, Kilian Lieret, Shunyu Yao, Karthik Narasimhan, and Ofir Press, SWE-agent: Agent-computer interfaces enable automated software engineering, in Proceedings of the 38th International Conference on Neural Information Processing Systems, NIPS ’24 (Curran Associates Inc, Red Hook, NY, 2025), ISBN [Amazon][WorldCat], https://proceedings.neurips.cc/paper_files/paper/2024/file/5a7c947568c1b1328ccc5230172e1e7c-Paper-Conference.pdf.
  17. Chris Lu, Cong Lu, Robert Tjarko Lange, Jakob Foerster, Jeff Clune, and David Ha, The AI scientist: Towards fully automated open-ended scientific discovery, arXiv:2408.06292.
  18. Daniil A. Boiko, Robert MacKnight, Ben Kline, and Gabe Gomes, Autonomous chemical research with large language models, Nature (London) 624, 570 (2023).
  19. Andres M. Bran, Sam Cox, Oliver Schilter, Carlo Baldassari, Andrew D. White, and Philippe Schwaller, Augmenting large language models with chemistry tools, Nat. Mach. Intell. 6, 525 (2024).
  20. Yixiang Ruan, Chenyin Lu, Ning Xu, Yuchen He, Yixin Chen, Jian Zhang, Jun Xuan, Jianzhang Pan, Qun Fang, Hanyu Gao et al., An automatic end-to-end chemical synthesis development platform powered by large language models, Nat. Commun. 15, 10160 (2024).
  21. Yusuf H. Roohani, Andrew H. Lee, Qian Huang, Jian Vora, Zachary Steinhart, Kexin Huang, Alexander Marson, Percy Liang, and Jure Leskovec, Biodiscoveryagent: An AI agent for designing genetic perturbation experiments, in Proceedings of the Thirteenth International Conference on Learning Representations (2025), https://openreview.net/forum?id=HAwZGLcye3.
  22. Yuanhao Qu, Kaixuan Huang, Ming Yin, Kanghong Zhan, Dyllan Liu, Di Yin, Henry C. Cousins, William A. Johnson, Xiaotong Wang, Mihir Shah et al., Crispr-GPT for agentic automation of gene-editing experiments, Nat. Biomed. Eng. 10, 245 (2025).
  23. Haining Pan, Nayantara Mudur, William Taranto, Maria Tikhanovskaya, Subhashini Venugopalan, Yasaman Bahri, Michael P. Brenner, and Eun-Ah Kim, Quantum many-body physics calculations with large language models, Commun. Phys. 8, 49 (2025).
  24. Tuan Dung Nguyen, Yuan-Sen Ting, Ioana Ciucă, Charlie O’Neill, Ze-Chang Sun, Maja Jabłońska, Sandor Kruk, Ernest Perkowski, Jack Miller, Jason Li et al., AstroLLaMA: Towards specialized foundation models in astronomy, arXiv:2309.06126.
  25. Xiaoxuan Wang, Ziniu Hu, Pan Lu, Yanqiao Zhu, Jieyu Zhang, Satyen Subramaniam, Arjun R. Loomba, Shichang Zhang, Yizhou Sun, and Wei Wang, SciBench: Evaluating college-level scientific problem-solving abilities of large language models, arXiv:2307.10635.
  26. Shi Qiu, Shaoyang Guo, Zhuo-Yang Song, Yunbo Sun, Zeyu Cai, Jiashen Wei, Tianyu Luo, Yixuan Yin, Haoxu Zhang, Yi Hu et al., PHYBench: Holistic evaluation of physical perception and reasoning in large language models, arXiv:2504.16074.
  27. Xinyu Zhang, Yuxuan Dong, Yanrui Wu, Jiaxing Huang, Chengyou Jia, Basura Fernando, Mike Zheng Shou, Lingling Zhang, and Jun Liu, Physreason: A comprehensive benchmark towards physics-based reasoning, arXiv:2502.12054.
  28. Minyang Tian, Luyu Gao, Shizhuo Dylan Zhang, Xinan Chen, Cunwei Fan, Xuefei Guo, Roland Haas, Pan Ji, Kittithat Krongchon et al., SciCode: A research coding benchmark curated by scientists, in Advances in Neural Information Processing Systems, edited by A. Globerson, L. Mackey, D. Belgrave, A. Fan, U. Paquet, J. Tomczak, and C. Zhang (Curran Associates, Inc., Red Hook, New York, 2024), Vol. 37, pp. 30624–30650, https://proceedings.neurips.cc/paper_files/paper/2024/file/36850592258c8c41cecdaa3dea5ff7de-Paper-Datasets_and_Benchmarks_Track.pdf.
  29. Sirui Lu, Zhijing Jin, Terry Jingchen Zhang, Pavel Kos, J Ignacio Cirac, and Bernhard Schölkopf, Can theoretical physics research benefit from language agents?, arXiv:2506.06214.
  30. Kristian G. Barman, Sascha Caron, Emily Sullivan, Henk W. de Regt, Roberto Ruiz de Austri, Mieke Boon, Michael Färber, Stefan Fröse, Faegheh Hasibi, Andreas Ipp et al., Large physics models: Towards a collaborative approach with large language models and foundation models, arXiv:2501.05382.
  31. Ziqi Wang, Hongshuo Huang, Hancheng Zhao, Changwen Xu, Shang Zhu, Jan Janssen, and Venkatasubramanian Viswanathan, DREAMS: Density functional theory based research engine for agentic materials simulation, arXiv:2507.14267.
  32. Licong Xu, Milind Sarkar, Anto I. Lonappan, Íñigo Zubeldia, Pablo Villanueva-Domingo, Santiago Casas, Christian Fidler, Chetana Amancharla, Ujjwal Tiwari, Adrian Bayer et al., Open source planning & control system with language agents for autonomous scientific discovery, arXiv:2507.07257.
  33. Santiago Casas, Christian Fidler, Boris Bolliet, Francisco Villaescusa-Navarro, and Julien Lesgourgues, CLAPP: The class LLM agent for pair programming, arXiv:2508.05728.
  34. Baohua Zhang, Xin Li, Huangchao Xu, Zhong Jin, Quansheng Wu, and Ce Li, Topomas: Large language model driven topological materials multiagent system, arXiv:2507.04053.
  35. Darui Lu, Jordan M. Malof, and Willie J. Padilla, An agentic framework for autonomous metamaterial modeling and inverse design, arXiv:2506.06935.
  36. Ankita Sharma, YuQi Fu, Vahid Ansari, Rishabh Iyer, Fiona Kuang, Kashish Mistry, Raisa Islam Aishy, Sara Ahmad, Joaquin Matres, Dirk R. Englund, and Joyce K. S. Poon, AI agents for photonic integrated circuit design automation, arXiv:2508.14123.
  37. Rong Sha, Binglin Wang, Jun Yang, Xiaoxiao Ma, Chengkun Wu, Liang Yan, Chao Zhou, Jixun Liu, Guochao Wang, Shuhua Yan, and Lingxiao Zhu, LLM-based multi-agent copilot for quantum sensor, arXiv:2508.05421.
  38. Shuxiang Cao, Zijian Zhang, Mohammed Alghadeer, Simone D. Fasciati, Michele Piscitelli, Mustafa Bakr, Peter Leek, and Alán Aspuru-Guzik, Automating quantum computing laboratory experiments with an agent-based AI framework, Patterns 6, 101372 (2025).
  39. OpenAI, GPT-5 system card, 2025, https://cdn.openai.com/gpt-5-system-card.pdf.
  40. See Supplemental Material at http://link.aps.org/supplemental/10.1103/xnqc-q6nt for details on analysis tools, experiment tools, physical systems, models discovered by the agent, and summaries of example explorations.
  41. Steven L. Brunton, Joshua L. Proctor, and J. Nathan Kutz, Discovering governing equations from data by sparse identification of nonlinear dynamical systems, Proc. Natl. Acad. Sci. U.S.A. 113, 3932 (2016).
  42. Google AI, Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities, arXiv:2507.06261.
  43. Mario Krenn, Jonas Landgraf, Thomas Foesel, and Florian Marquardt, Artificial intelligence and machine learning for quantum technologies, Phys. Rev. A 107, 010101 (2023).
  44. Agnes Valenti, Evert van Nieuwenburg, Sebastian Huber, and Eliska Greplova, Hamiltonian learning for quantum error correction, Phys. Rev. Res. 1, 033092 (2019).
  45. Mikel Landajuela, Chak Shing Lee, Jiachen Yang, Ruben Glatt, Claudio P. Santiago, Ignacio Aravena, Terrell Mundhenk, Garrett Mulcahy, and Brenden K. Petersen, A unified framework for deep symbolic regression, Proceedings of the 36th International Conference on Neural Information Processing Systems, NIPS ’22 (2022), Vol. 35, pp. 33985–33998, https://proceedings.neurips.cc/paper_files/paper/2022/file/dbca58f35bddc6e4003b2dd80e42f838-Paper-Conference.pdf.
  46. Brian de Silva, Kathleen Champion, Markus Quade, Jean-Christophe Loiseau, J. Kutz, and Steven Brunton, pysindy: A python package for the sparse identification of nonlinear dynamical systems from data, J. Open Source Software 5, 2104 (2020).
  47. Samuel H. Rudy, Steven L. Brunton, Joshua L. Proctor, and J. Nathan Kutz, Data-driven discovery of partial differential equations, Sci. Adv. 3, e1602614 (2017).
  48. Silviu-Marian Udrescu and Max Tegmark, AI Feynman: A physics-inspired method for symbolic regression, Sci. Adv. 6, eaay2631 (2020).
  49. Silviu-Marian Udrescu, Andrew Tan, Jiahai Feng, Orisvaldo Neto, Tailin Wu, and Max Tegmark, AI Feynman 2.0: Pareto-optimal symbolic regression exploiting graph modularity, Advances in Neural Information Processing Systems (2020), Vol. 33, pp. 4860–4871, https://proceedings.neurips.cc/paper/2020/hash/33a854e247155d590883b93bca53848a-Abstract.html.
  50. Kevin Maik Jablonka, Philippe Schwaller, Andres Ortega-Guerrero, and Berend Smit, Leveraging large language models for predictive chemistry, Nat. Mach. Intell. 6, 161 (2024).
  51. Joren Van Herck, María Victoria Gil, Kevin Maik Jablonka, Alex Abrudan, Andy S. Anker, Mehrdad Asgari, Ben Blaiszik, Antonio Buffo, Leander Choudhury, Clemence Corminboeuf et al., Assessment of fine-tuned large language models for real-world chemistry and material science applications, Chem. Sci. 16, 670 (2025).
  52. Jinhyuk Lee, Wonjin Yoon, Sungdong Kim, Donghyeon Kim, Sunkyu Kim, Chan Ho So, and Jaewoo Kang, Biobert: A pre-trained biomedical language representation model for biomedical text mining, Bioinformatics 36, 1234 (2019).
  53. Renqian Luo, Liai Sun, Yingce Xia, Tao Qin, Sheng Zhang, Hoifung Poon, and Tie-Yan Liu, Biogpt: Generative pre-trained transformer for biomedical text generation and mining, Briefings Bioinf. 23, bbac409 (2022).
  54. Nägele Maximilian and Marquardt Florian, sciexplorer framework, https://github.com/MaxNaeg/SciExplorer.git (2025).
  55. Nägele Maximilian and Marquardt Florian, sciexplorer results, https://github.com/MaxNaeg/SciExplorerResults.git (2025).
  56. James Bradbury, Roy Frostig, Peter Hawkins, Matthew James Johnson, Chris Leary, Dougal Maclaurin, George Necula, Adam Paszke, Jake VanderPlas, Skye Wanderman-Milne, and Qiao Zhang, jax: Composable transformations of Python+NumPy programs, http://github.com/jax-ml/jax (2018).

Outline

Information

Phys. Rev. X 16, 031002– Published 8 July, 2026
  • Received 12 November 2025
  • Accepted 22 May 2026
Crossmark - Check for updates button
Creative Commons Logo

Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI. Open access publication funded by the Max Planck Society.

Published by the American Physical Society

Sign In to Your Journals Account

Filter

Filter

Article Lookup

Enter a citation