arXiv subjects

Mathew J. Cherukara

Publications and source records attributed to Mathew J. Cherukara.

At least 19 recordsLinked to original sources

arXiv2026

APEXA: Execution-Integrity Enforcement for Multi-Agent LLM Automation of Synchrotron Data Reduction

Synchrotron data reduction, detector calibration followed by azimuthal integration of terabyte-scale diffraction series, is a multi-step, expert-bound bottleneck that increasingly limits the science rate of user facilities. LLM agents promise to collapse it, but driving a real pipeline with a stochastic model creates a failure mode chat benchmarks cannot see: an agent can report a calibration that was never computed. Correctness here is a property of what executed, not of the transcript. We present APEXA, a deployed multi-agent framework (61 tools over heterogeneous compute, run as a single reasoning loop) automating calibration and integration from natural language at a major light source. We make three contributions. First, execution-integrity enforcement: a deterministic tool-layer guard that refuses to surface any result not backed by an executed tool call, with a parser tolerant of cross-model tool-call format drift: in deployment, a frontier model fabricated a complete calibration-comparison report for commands that never ran, which the guard converts to an explicit non-result; the same code gates an optional motor-control surface at 0/200 adversarial violations against a simulated IOC, versus 15/200 for an equivalent safety prompt. Second, we release APEXA-Bench, an evaluation harness of 58 facility tasks (50 base plus an 8-task cross-detector slice) organized by a four-class physical-consequence taxonomy, the first benchmark axis we know of separating a wasted compute cycle from a damaged instrument; its cross-detector grading against NIST-traceable lattice constants surfaced two latent pipeline bugs. Large-scale agent scoring is left to a full-length study. Third, we validate APEXA on real beamline data: from one natural-language prompt it recovers detector geometry and integrates a full attenuation/exposure sweep. We release the framework, harness and traces.

cs.AI↗

arXiv2026

Acquisition Geometry-Assisted Whole-Group Localization of X-ray Fluorescence Maps in Optical Microscopy Images

X-ray fluorescence (XRF) microscopy maps elemental distributions, while optical microscopy can provide complementary morphological context. Localizing XRF fields of view (FOVs) in optical images is difficult because the two modalities differ in contrast mechanism and resolution. Most current workflows place each XRF tile independently, even when acquisition metadata already record the tiles' relative scan positions. This study formalizes XRF tile-group localization, in which one optical-frame placement is estimated for the whole group, constrained by acquisition geometry and quantified using group intersection-over-union (GroupIoU). In a controlled case study, independent localization failed with GroupIoU 0.000, whereas group localization achieved 0.931. Replacing the normalized cross-correlation (NCC) metric with mutual information (MI) gave nearly identical results, showing that the outcome is not specific to one local similarity metric. In another multiscale case study, using a coarse XRF survey scan to connect the fine-scale tile group to the optical image increased mean GroupIoU from 0.694 to 0.856. These case studies support using acquisition geometry as an explicit constraint when localizing related XRF tiles.

cs.CV↗

arXiv2026

XRF-to-Optical Field-of-View Localization with Vision Language Models

Registering images acquired with different microscopy modalities is essential for relating complementary measurements of the same specimen. In correlative X-ray fluorescence (XRF) and optical microscopy, the XRF map often covers only a small region of an optical image acquired from the same or an adjacent tissue section. Field-of-view (FOV) localization is necessary but can be difficult when appearance and structure differ across modalities. Here we evaluate training-free vision language model (VLM) localization on two datasets representing same-section high-correspondence and adjacent-section low-correspondence imaging. We test unconstrained and metadata-constrained search and compare VLMs with geometric controls, classical template matching, and two alternative training-free approaches (DINOv2 and multiGradICON). Direct VLM prompting produced content-dependent spatial signals but was not reliable alone. Classical matching was most accurate when cross-modal structure was preserved but failed in the low-correspondence collection. A proposal-and-verify workflow used repeated VLM predictions as candidates and image-based similarity to select the final location. This workflow recovered useful localization in the low-correspondence regime.

cs.CV↗

arXiv2026

Artifact-centered Claim-aware Observability for Autonomous Scientific Agents

Autonomous scientific agents now increasingly propose ideas, write code, run experiments, analyze results, and even draft papers. Observe and audit those agents are necessary but logging every model call is not enough, scientists also need to inspect the artifacts and claims that the systems produced and their relations. This is driven by the fact that failures in scientific agent systems are often distributed across several objects. A manuscript claim may cite the wrong evidence, a search process may select a degenerate candidate, a laboratory novelty claim may depend on an unstated rule, or a multi-agent plan may change without a visible trigger. Existing tracing, experiment tracking, and archival provenance tools are valuable, but their native objects do not make these scientific audit relations first-class. We argue that autonomous scientific systems should emit portable, claim-aware artifact lineage as a minimum audit layer. We propose a compact observability profile organized around individuals, operators, fitness records, lineage, archives, runs, streams, and steering commands. In this profile, scientific claims are ordinary individuals with explicit evidence bindings and verification records. The profile is intended as a semantic layer that complements current telemetry and provenance standards. Execution details can remain in OpenTelemetry. Final packages can export to PROV-O or RO-Crate standards.

cs.CL↗

arXiv2026

A corrective agentic hybrid RAG and an operations-grounded evaluation for a scientific facility

Scientific user facilities accumulate decades of operational knowledge that no single search index covers: electronic logbooks, technical documents, internal wikis, operations chat messages, maintenance records, and live control-system data. We present APS-RAG, Advanced Photon Source Retrieval Augmented Generation, a deployed platform that makes the institutional knowledge at the Advanced Photon Source (APS) accessible to staff through natural-language queries, along with an operations-grounded evaluation. The retrieval engine fuses dense, sparse, and knowledge-graph (KG) channels with query-type-adaptive reciprocal-rank fusion, adds a corrective agentic loop, and runs a native-tool ReAct executor over a Model Context Protocol (MCP) tooling layer. We construct APS-Bench, a 50-question, question-answering (QA) dataset with auditable gold answers. Every retrieval-augmented variant numerically improves strict vital-nugget recall over a naive BM25 baseline (63.8%), with the full corrective Agentic GraphRAG scoring (70.3%). The cross-encoder reranker contributes significantly to answer quality: removing it and allowing the LLM to score relevance drastically reduces strict vital recall by 32.8%. The graph channel and corrective loop contribute positively as expected, but the performance gains are marginal. Additionally, we also compare the performance of open-source and closed-source LLMs in final answer synthesis. We release the APS-Bench construction methodology, the six-layer evaluation harness, and the underlying codebase, along with the '/aps-rag' retrieval agent skill framework, to support reproduction and adoption at other facilities. Together, the deployed platform and its operations-grounded evaluation present a promising workflow for trustworthy, statistically grounded AI assistance in facility operations, transferable to other large scientific instruments.

physics.acc-ph↗

arXiv2026

CVEvolve: Autonomous Algorithm Discovery for Unstructured Scientific Data Processing

Scientific data processing often requires task-specific algorithms or AI models, creating a barrier for domain scientists who need to analyze their data but may not have extensive computing or image-processing expertise. This barrier is especially pronounced when data are noisy, have a high dynamic range, are sparsely labeled, or are only loosely specified. We introduce CVEvolve, an autonomous agentic harness with a zero-code interface for scientific data-processing algorithm discovery. CVEvolve combines a multi-round search strategy with tools for code execution, evaluation implementation, history management, holdout testing, and optional inspection of scientific data and visual outputs. The search alternates between discovery and improvement actions, and uses lineage-aware stochastic candidate sampling to balance exploration and exploitation. We demonstrate CVEvolve on X-ray fluorescence microscopy image registration, Bragg peak detection, high-energy diffraction microscopy image segmentation, and hybrid analytical-learning-based affine registration. Across these tasks, CVEvolve discovers algorithms that improve over baseline methods, while holdout test tracking helps identify candidates that generalize better than later over-optimized alternatives. These results show that zero-code, autonomous LLM-powered algorithm development can help domain scientists turn unstructured scientific image data into practical algorithms and downstream scientific discoveries.

cs.AI↗

arXiv2026

AlphaDiffract: Automated Crystallographic Analysis of Powder X-ray Diffraction Data

Materials identification and structural understanding from powder X-ray diffraction (PXRD) data is a long-standing challenge in materials science, fundamental to discovering and characterizing novel materials. A prerequisite for full structure solution is the accurate determination of the crystal lattice, including lattice parameters and crystallographic symmetries. Traditional methods for this are iterative and typically require expert input, and while existing deep learning approaches have shown promise, a robust, single-shot method for comprehensive lattice determination from experimental data remains a key goal. Here, we introduce AlphaDiffract, a deep learning framework that achieves state-of-the-art performance in predicting the crystal system, space group, and lattice parameters directly from PXRD patterns. AlphaDiffract utilizes a 1D adaptation of the ConvNeXt architecture, a modern convolutional neural network that integrates key design principles from transformers, coupled with dedicated prediction heads for each crystallographic property. The model is trained on the largest-to-date physics-based dataset of over 31 million simulated diffraction patterns, generated by augmenting 312,267 curated structures from the ICSD and Materials Project databases. Crucially, it demonstrates strong generalization to experimental data, achieving 81.7% crystal system accuracy and 66.2% space group accuracy on the RRUFF dataset while additionally predicting all six lattice parameters. By providing a unified model for rapid and accurate lattice determination from PXRD data, AlphaDiffract represents a significant step forward in leveraging deep learning for high-throughput materials discovery.

cond-mat.mtrl-sci↗

arXiv2026

EAA: Automating materials characterization with vision language model agents

We present Experiment Automation Agents (EAA), a vision-language-model-driven agentic system designed to automate complex experimental microscopy workflows. EAA integrates multimodal reasoning, tool-augmented action, and optional long-term memory to support both autonomous procedures and interactive user-guided measurements. Built on a flexible task-manager architecture, the system enables workflows ranging from fully agent-driven automation to logic-defined routines that embed localized LLM queries. EAA further provides a modern tool ecosystem with two-way compatibility for Model Context Protocol (MCP), allowing instrument-control tools to be consumed or served across applications. We demonstrate EAA at an imaging beamline at the Advanced Photon Source, including automated zone plate focusing, natural language-described feature search, and interactive data acquisition. These results illustrate how vision-capable agents can enhance beamline efficiency, reduce operational burden, and lower the expertise barrier for users.

cs.AI↗

arXiv2025

Physically Interpretable Interatomic Potentials via Symbolic Regression and Reinforcement Learning

The development of next-generation molecular simulation models requires moving beyond pre-defined functional forms toward machine learning (ML) techniques that directly capture multiscale physics. Here, we demonstrate such an approach using symbolic regression (SR) with equation learner networks and a reinforcement learning search engine to derive interpretable equations for interatomic interactions. Training data were generated through nested ensemble sampling with density functional theory (DFT) energetics, spanning crystalline to highly disordered states. The optimization of the learner network employed continuous-action Monte Carlo Tree Search (MCTS) combined with gradient descent, enabling efficient exploration of function space. For copper as a representative transition metal, an unconstrained search produced models that outperformed fixed-form Sutton-Chen EAM potentials. The SR-derived models (SR1 and SR2) reproduced key material properties - lattice constants, cohesive energies, equations of state, elastic constants, phonon dispersion, defect formation energies, surface/bulk energetics, and phase transformation with significantly improved accuracy. Furthermore, stringent melting simulations using two-phase solid-amorphous interfaces confirmed that SR models accurately capture the interplay of vibrational entropy, cohesive energy, and structural dynamics, surpassing SC-EAM in both qualitative and quantitative predictions. This highlights the potential of SR to deliver fast, accurate, flexible, and physically meaningful potentials, advancing predictive modeling across scales.

cond-mat.mtrl-sci↗

arXiv2025

Fidelity-preserving enhancement of ptychography with foundational text-to-image models

Ptychographic phase retrieval enables high-resolution imaging of complex samples but often suffers from artifacts such as grid pathology and multislice crosstalk, which degrade reconstructed images. We propose a plug-and-play (PnP) framework that integrates physics model-based phase retrieval with text-guided image editing using foundational diffusion models. By employing the alternating direction method of multipliers (ADMM), our approach ensures consensus between data fidelity and artifact removal subproblems, maintaining physics consistency while enhancing image quality. Artifact removal is achieved using a text-guided diffusion image editing method (LEDITS++) with a pre-trained foundational diffusion model, allowing users to specify artifacts for removal in natural language. Demonstrations on simulated and experimental datasets show significant improvements in artifact suppression and structural fidelity, validated by metrics such as peak signal-to-noise ratio (PSNR) and diffraction pattern consistency. This work highlights the combination of text-guided generative models and model-based phase retrieval algorithms as a transferable and fidelity-preserving method for high-quality diffraction imaging.

cs.GR↗

arXiv2025

Operating advanced scientific instruments with AI agents that learn on the job

Advanced scientific user facilities, such as next generation X-ray light sources and self-driving laboratories, are revolutionizing scientific discovery by automating routine tasks and enabling rapid experimentation and characterizations. However, these facilities must continuously evolve to support new experimental workflows, adapt to diverse user projects, and meet growing demands for more intricate instruments and experiments. This continuous development introduces significant operational complexity, necessitating a focus on usability, reproducibility, and intuitive human-instrument interaction. In this work, we explore the integration of agentic AI, powered by Large Language Models (LLMs), as a transformative tool to achieve this goal. We present our approach to developing a human-in-the-loop pipeline for operating advanced instruments including an X-ray nanoprobe beamline and an autonomous robotic station dedicated to the design and characterization of materials. Specifically, we evaluate the potential of various LLMs as trainable scientific assistants for orchestrating complex, multi-task workflows, which also include multimodal data, optimizing their performance through optional human input and iterative learning. We demonstrate the ability of AI agents to bridge the gap between advanced automation and user-friendly operation, paving the way for more adaptable and intelligent scientific facilities.

physics.ins-det↗

arXiv2025

DONUT: Physics-aware Machine Learning for Real-time X-ray Nanodiffraction Analysis

Coherent X-ray scattering techniques are critical for investigating the fundamental structural properties of materials at the nanoscale. While advancements have made these experiments more accessible, real-time analysis remains a significant bottleneck, often hindered by artifacts and computational demands. In scanning X-ray nanodiffraction microscopy, which is widely used to spatially resolve structural heterogeneities, this challenge is compounded by the convolution of the divergent beam with the sample's local structure. To address this, we introduce DONUT (Diffraction with Optics for Nanobeam by Unsupervised Training), a physics-aware neural network designed for the rapid and automated analysis of nanobeam diffraction data. By incorporating a differentiable geometric diffraction model directly into its architecture, DONUT learns to predict crystal lattice strain and orientation in real-time. Crucially, this is achieved without reliance on labeled datasets or pre-training, overcoming a fundamental limitation for supervised machine learning in X-ray science. We demonstrate experimentally that DONUT accurately extracts all features within the data over 200 times more efficiently than conventional fitting methods.

cs.LG↗

arXiv2025

Ptychographic Image Reconstruction from Limited Data via Score-Based Diffusion Models with Physics-Guidance

Ptychography is a data-intensive computational imaging technique that achieves high spatial resolution over large fields of view. The technique involves scanning a coherent beam across overlapping regions and recording diffraction patterns. Conventional reconstruction algorithms require substantial overlap, increasing data volume and experimental time, reaching PiB-scale experimental data and weeks to month-long data acquisition times. To address this, we propose a reconstruction method employing a physics-guided score-based diffusion model. Our approach trains a diffusion model on representative object images to learn an object distribution prior. During reconstruction, we modify the reverse diffusion process to enforce data consistency, guiding reverse diffusion toward a physically plausible solution. This method requires a single pretraining phase, allowing it to generalize across varying scan overlap ratios and positions. Our results demonstrate that the proposed method achieves high-fidelity reconstructions with only a 20% overlap, while the widely employed rPIE method requires a 62% overlap to achieve similar accuracy. This represents a significant reduction in data requirements, offering an alternative to conventional techniques.

eess.IV↗

arXiv2025

Demonstration of an AI-driven workflow for dynamic x-ray spectroscopy

X-ray absorption near edge structure (XANES) spectroscopy is a powerful technique for characterizing the chemical state and symmetry of individual elements within materials, but requires collecting data at many energy points which can be time-consuming. While adaptive sampling methods exist for efficiently collecting spectroscopic data, they often lack domain-specific knowledge about XANES spectra structure. Here we demonstrate a knowledge-injected Bayesian optimization approach for adaptive XANES data collection that incorporates understanding of spectral features like absorption edges and pre-edge peaks. We show this method accurately reconstructs the absorption edge of XANES spectra using only 15-20% of the measurement points typically needed for conventional sampling, while maintaining the ability to determine the x-ray energy of the sharp peak after absorption edge with errors less than 0.03 eV, the absorption edge with errors less than 0.1 eV; and overall root-mean-square errors less than 0.005 compared to compared to traditionally sampled spectra. Our experiments on battery materials and catalysts demonstrate the method's effectiveness for both static and dynamic XANES measurements, improving data collection efficiency and enabling better time resolution for tracking chemical changes. This approach advances the degree of automation in XANES experiments reducing the common errors of under- or over-sampling points in near the absorption edge and enabling dynamic experiments that require high temporal resolution or limited measurement time.

physics.app-ph↗

arXiv2024

Data-driven discovery of dynamics from time-resolved coherent scattering

Coherent X-ray scattering (CXS) techniques are capable of interrogating dynamics of nano- to mesoscale materials systems at time scales spanning several orders of magnitude. However, obtaining accurate theoretical descriptions of complex dynamics is often limited by one or more factors -- the ability to visualize dynamics in real space, computational cost of high-fidelity simulations, and effectiveness of approximate or phenomenological models. In this work, we develop a data-driven framework to uncover mechanistic models of dynamics directly from time-resolved CXS measurements without solving the phase reconstruction problem for the entire time series of diffraction patterns. Our approach uses neural differential equations to parameterize unknown real-space dynamics and implements a computational scattering forward model to relate real-space predictions to reciprocal-space observations. This method is shown to recover the dynamics of several computational model systems under various simulated conditions of measurement resolution and noise. Moreover, the trained model enables estimation of long-term dynamics well beyond the maximum observation time, which can be used to inform and refine experimental parameters in practice. Finally, we demonstrate an experimental proof-of-concept by applying our framework to recover the probe trajectory from a ptychographic scan. Our proposed framework bridges the wide existing gap between approximate models and complex data.

cond-mat.mtrl-sci↗

arXiv2024

Deep Learning of Structural Morphology Imaged by Scanning X-ray Diffraction Microscopy

Scanning X-ray nanodiffraction microscopy is a powerful technique for spatially resolving nanoscale structural morphologies by diffraction contrast. One of the critical challenges in experimental nanodiffraction data analysis is posed by the convergence angle of nanoscale focusing optics which creates simultaneous dependency of the far-field scattering data on three independent components of the local strain tensor - corresponding to dilation and two potential rigid body rotations of the unit cell. All three components are in principle resolvable through a spatially mapped sample tilt series however traditional data analysis is computationally expensive and prone to artifacts. In this study, we implement NanobeamNN, a convolutional neural network specifically tailored to the analysis of scanning probe X-ray microscopy data. NanobeamNN learns lattice strain and rotation angles from simulated diffraction of a focused X-ray nanobeam by an epitaxial thin film and can directly make reasonable predictions on experimental data without the need for additional fine-tuning. We demonstrate that this approach represents a significant advancement in computational speed over conventional methods, as well as a potential improvement in accuracy over the current standard.

physics.app-ph↗

arXiv2024

Predicting ptychography probe positions using single-shot phase retrieval neural network

Ptychography is a powerful imaging technique that is used in a variety of fields, including materials science, biology, and nanotechnology. However, the accuracy of the reconstructed ptychography image is highly dependent on the accuracy of the recorded probe positions which often contain errors. These errors are typically corrected jointly with phase retrieval through numerical optimization approaches. When the error accumulates along the scan path or when the error magnitude is large, these approaches may not converge with satisfactory result. We propose a fundamentally new approach for ptychography probe position prediction for data with large position errors, where a neural network is used to make single-shot phase retrieval on individual diffraction patterns, yielding the object image at each scan point. The pairwise offsets among these images are then found using a robust image registration method, and the results are combined to yield the complete scan path by constructing and solving a linear equation. We show that our method can achieve good position prediction accuracy for data with large and accumulating errors on the order of $10^2$ pixels, a magnitude that often makes optimization-based algorithms fail to converge. For ptychography instruments without sophisticated position control equipment such as interferometers, our method is of significant practical potential.

physics.app-ph↗

arXiv2024

Three-dimensional Hard X-ray Ptychographic Reflectometry Imaging on Extended Mesoscopic Surface Structures

Many nano and quantum devices, with their sizes often spanning from millimeters down to sub-nanometer, have intricate low-dimensional, non-uniform, or hierarchical structures on surfaces and interfaces. Since their functionalities are dependent on these structures, high-resolution surface-sensitive characterization becomes imperative to gain a comprehensive understanding of the function-structure relationship. We thus developed hard X-ray ptychographic reflectometry imaging, a new technique that merges the high-resolution two-dimensional imaging capabilities of hard X-ray ptychography for extended objects, with the high-resolution depth profiling capabilities of X-ray reflectivity for layered structures. The synergy of these two methods fully leverages both amplitude and phase information from ptychography reconstruction to not only reveal surface topography and localized structures such as shapes and electron densities, but also yields statistical details such as interfacial roughness that is not readily accessible through coherent imaging solely. The hard X-ray ptychographic reflectometry imaging is well-suited for three-dimensional imaging of mesoscopic samples, particularly those comprising planar or layered nanostructures on opaque supports, and could also offer a high-resolution surface metrology and defect analysis on semiconductor devices such as integrated nanocircuits and lithographic photomasks for microchip fabrications.

physics.optics↗