FUSION is a skill‑based research agent that automates the use of publicly available nuclear‑physics codes by retrieving the code, running verified inputs, parsing outputs, and ensuring results meet specified benchmarks. The current release supports twenty different codes across various nuclear physics domains and includes a searchable collection of literature references. Documentation and source code are freely available under an MIT license.
arXiv:2602. 10132v3 Announce Type: replace-cross Abstract: Development and operation of commercially viable fusion energy reactors such as tokamaks require accurate predictions of plasma dynamics from sparse, noisy, and incomplete sensors readings.
arXiv:2608. 12249v1 Announce Type: new Abstract: Modernizing legacy Fortran is a problem of volume: the transformations are individually routine, but the codebases can be enormous, and across much of computational science the work simply goes undone.
VerifiedLab is a multi‑agent framework that transforms research records into active systems for laboratory experiments. It employs a team of AI scientists—principal investigator, librarian, specialists, gate, and critic—to reason over records field by field, verify claims against measured reliability, and decide whether to release or withhold information. Experiments on perovskite solar cells show that VerifiedLab can recover known physical trends, anticipate future device reports, and provide a trust metric for inherited literature.
arXiv:2607. 09789v1 Announce Type: new Abstract: We introduce PHITSBench, an execution-scored benchmark for the Monte Carlo Particle and Heavy Ion Transport code System (PHITS).
arXiv:2606. 07550v1 Announce Type: cross Abstract: Offline reinforcement learning (RL) offers a promising route for developing plasma controllers from historical tokamak data, since online trial-and-error on real devices is costly and risky.
arXiv:2604. 01313v2 Announce Type: replace Abstract: High-fidelity simulations and complex inverse problems, such as detector modeling and unfolding, are computationally intensive bottlenecks across subatomic physics, yet essential for accurate physical interpretation.
The paper introduces Scientific Agent Skills, an open library of 163 procedural knowledge modules across 16 scientific domains such as genomics, cheminformatics, medical imaging, study design, and scientific communication. Each skill is organized as a directory containing a versioned, human‑readable instruction file, reference material, and runnable scripts, allowing a language‑model agent to load the appropriate procedure only when needed. The library is publicly licensed and hosted on GitHub, but the authors report no task‑level evaluation or host selection rate.
arXiv:2607. 10366v1 Announce Type: new Abstract: Offline-trained surrogates for Inertial Confinement Fusion (ICF) suffer a well-known failure mode that iterative optimizers drive inputs into out-of-distribution (OOD) regions where predictions become unreliable.
arXiv:2509. 17255v2 Announce Type: replace-cross Abstract: We present the first language-model-driven agentic artificial intelligence (AI) system to autonomously execute multi-stage physics experiments on a production synchrotron light source.
IGNITE is a generative world foundation model designed to simulate fusion plasma behavior in the DIII‑D National Fusion Facility. Trained in a self‑supervised way on more than a decade of unlabeled experimental data, its core dynamics model can predict entire discharges from specified actuator trajectories, which may be supplied or generated from textual prompts or desired outcomes. The architecture uses spatio‑temporal tokenizers for diverse input modalities—time‑series measurements, image sequences, and high‑resolution spectrograms—feeding an auto‑regressive backbone that can forecast discharges over an infinite horizon.
arXiv:2610.32441v1 Announce Type: new Abstract: Large language model (LLM) agents are increasingly capable of supporting scientific workflows, but their capabilities across diverse research activitie...
The paper presents a reproducible workflow built in the Galaxy environment for automated crack identification and damage assessment from scanning electron microscopy images of fusion materials. It processes SEM images and metadata to generate crack masks, skeletonized networks, quality‑control visualizations, and scalar damage descriptors without needing image‑specific tuning across different tungsten grades, microstructures, magnifications, and damage states. The workflow was demonstrated on a dataset of 418 images from 114 experiments covering five tungsten grades and three microstructural states, and it includes a crack‑density descriptor that serves as a standardized input for downstream machine‑learning prediction and physics‑based crack simulation modules.