Unmodeled states and uncertain action outcomes in agentic scanning tunneling microscopy
Abstract
In physical experiments, interventions can alter hidden experimental states in ways that cannot be predicted in advance. Autonomous scientific agents must therefore interpret the consequences of their interventions while operating with incomplete knowledge of the experimental state. Here we investigate this problem using scanning tunneling microscopy (STM) tip conditioning, traditionally a human-expert-intensive task governed by inaccessible tip apex conditions and uncertain action outcomes. We introduce quailbot, an agent harness that places the LLM inside the instrument feedback loop by linking physical interventions with experimental readbacks. After a brief apprenticeship with a human expert, frontier LLM agents autonomously completed end-to-end tip conditioning on an STM and passed an independent verification. The contingencies that arose during the experiments further exposed the limits of LLM agentic autonomy when relevant experimental states or dynamics were hidden or unmodeled. Our results show that future autonomous physical experimentation requires agents to infer hidden and unmodeled experimental states, track the outcomes of their actions, and operate within the observability and action limits of real instruments.
Related
- Parametrized Relativistic Quantum Theory: Framework and Application to Scanning Tunneling Microscopy
- Physics-Grounded Simulation Enables Zero-Shot Molecular and Reaction Analysis in Scanning Tunneling Microscopy
- Electronic theory for scanning tunneling microscopy spectra in bilayer nickelate thin films
- Atomic‐Scale Surface Imaging of Bulk Epitaxial CsPbBr₃ Perovskite Single Crystals on Mica Using Light Assisted Scanning Tunneling Microscopy at Low‐Temperature (80 K)
- Sub-diffraction-resolved spatial distribution of emitting excitons in scanning tunneling microscopy-induced luminescence of two-dimensional semiconductors via Richardson-Lucy deconvolution
BibTeX
@misc{cheng2026unmodeled,
title = {{Unmodeled states and uncertain action outcomes in agentic scanning tunneling microscopy}},
author = {Siyu Cheng and Muxian Xu and Christopher Candelora and Ilija Zeljkovic},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2609.27302},
eprint = {2609.27302},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.27302},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Unmodeled states and uncertain action outcomes in agentic scanning tunneling microscopy
AU - Siyu Cheng
AU - Muxian Xu
AU - Christopher Candelora
AU - Ilija Zeljkovic
PY - 2026
DA - 2026/10/23
PB - arXiv
DO - 10.48550/arxiv.2609.27302
UR - https://arxiv.org/abs/2609.27302
AB - In physical experiments, interventions can alter hidden experimental states in ways that cannot be predicted in advance. Autonomous scientific agents must therefore interpret the consequences of their interventions while operating with incomplete knowledge of the experimental state. Here we investigate this problem using scanning tunneling microscopy (STM) tip conditioning, traditionally a human-expert-intensive task governed by inaccessible tip apex conditions and uncertain action outcomes. We introduce quailbot, an agent harness that places the LLM inside the instrument feedback loop by linking physical interventions with experimental readbacks. After a brief apprenticeship with a human expert, frontier LLM agents autonomously completed end-to-end tip conditioning on an STM and passed an independent verification. The contingencies that arose during the experiments further exposed the limits of LLM agentic autonomy when relevant experimental states or dynamics were hidden or unmodeled. Our results show that future autonomous physical experimentation requires agents to infer hidden and unmodeled experimental states, track the outcomes of their actions, and operate within the observability and action limits of real instruments.
ER -