We present a sensor-fusion framework for rapid, non-destructive classification and quality control of organic substances, built on a standard Arduino Mega 2560 microcontroller platform equipped with three commercial environmental and gas sensors. All data used in this study were generated in-house: sensor outputs for ten distinct classes - including fresh and expired samples of apple juice, onion, garlic, and ginger, as well as cinnamon and cardamom - were systematically collected and labeled using this hardware setup, resulting in a unique, application-specific dataset. Correlation analysis was employed as part of the preprocessing pipeline for feature selection. After preprocessing and dimensionality reduction (PCA/LDA), multiple supervised learning models - including Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF), each with hyperparameter tuning, as well as an Artificial Neural Network (ANN) and an ensemble voting classifier - were trained and cross-validated on the collected dataset. The best-performing models, including tuned Random Forest, ensemble, and ANN, achieved test accuracies in the 93 to 94 percent range. These results demonstrate that low-cost, multisensory platforms based on the Arduino Mega 2560, combined with advanced machine learning and correlation-driven feature engineering, enable reliable identification and quality control of organic compounds.

Large language models (LLMs) have enabled powerful advances in natural language understanding and generation. Yet their application to complex, real-world scientific workflows remain limited by challenges in memory, planning, and tool integration. Here, we introduce SciBORG (Scientific Bespoke Artificial Intelligence Agents Optimized for Research Goals), a modular agentic framework that allows LLM-based agents to autonomously plan, reason, and achieve robust and reliable domain-specific task execution. Agents are constructed dynamically from source code documentation and augmented with finite-state automata (FSA) memory, enabling persistent state tracking and context-aware decision-making. This approach eliminates the need for manual prompt engineering and allows for robust, scalable deployment across diverse applications via maintaining context across extended workflows and to recover from tool or execution failures. We validate SciBORG through integration with both physical and virtual hardware, such as microwave synthesizers for executing user-specified reactions, with context-aware decision making and demonstrate its use in autonomous multi-step bioassay retrieval from the PubChem database utilizing multi-step planning, reasoning, agent-to-agent communication and coordination for execution of exploratory tasks. Systematic benchmarking shows that SciBORG agents achieve reliable execution, adaptive planning, and interpretable state transitions. Our results show that memory and state awareness are critical enablers of agentic planning and reliability, offering a generalizable foundation for deploying AI agents in complex environments.

The accurate prediction of phase diagrams is of central importance for both the fundamental understanding of materials as well as for technological applications in material sciences. However, the computational prediction of the relative stability between phases based on their free energy is a daunting task, as traditional free energy estimators require a large amount of simulation data to obtain uncorrelated equilibrium samples over a grid of thermodynamic states. In this work, we develop deep generative machine learning models for entire phase diagrams, employing normalizing flows conditioned on the thermodynamic states, e.g., temperature and pressure, that they map to. By training a single normalizing flow to transform the equilibrium distribution sampled at only one reference thermodynamic state to a wide range of target temperatures and pressures, we can efficiently generate equilibrium samples across the entire phase diagram. Using a permutation-equivariant architecture allows us, thereby, to treat solid and liquid phases on the same footing. We demonstrate our approach by predicting the solid-liquid coexistence line for a Lennard-Jones system in excellent agreement with state-of-the-art free energy methods while significantly reducing the number of energy evaluations needed.

A student in Carnegie Mellon University's School of Computer Science (SCS) has used reinforcement learning to help control nuclear fusion reactions, a significant step toward harnessing the immense power produced in nuclear fusion as a source of clean, abundant energy. Ian Char, a doctoral candidate in the Machine Learning Department, used reinforcement learning to control the hydrogen plasma of the tokamak machine at the DIII-D National Fusion Facility in San Diego. He was the first CMU researcher to run an experiment on the sought-after machines, the first to use reinforcement learning to affect the rotation of a tokamak plasma, and the first person to try reinforcement learning on the largest operating tokamak machine in the United States. Char collaborated with the Princeton Plasma Physics Laboratory (PPPL) on the work. "Reinforcement learning affected the plasma's pressure and its rotation," Char said.

We propose a machine learning approach to predict the shapes of the longitudinal spin fluctuation (LSF) energy landscapes for each local magnetic moment. This approach allows the inclusion of the effects of LSFs in, e.g., the simulation of a magnetic material with ab initio molecular dynamics in an effective way. This type of simulation requires knowledge of the reciprocal interaction between atoms and moments, which, in principle, would entail calculating the energy landscape of each atom at every instant in time. The machine learning approach is based on the kernel ridge regression method and developed using bcc Fe at the Curie temperature and ambient pressure as a test case. We apply the trained machine learning models in a combined atomistic spin dynamics and ab initio molecular dynamics (ASD-AIMD) simulation, where they are used to determine the sizes of the magnetic moments of every atom at each time step. In addition to running an ASD-AIMD simulation with the LSF machine learning approach for bcc Fe at the Curie temperature, we also simulate Fe at temperature and pressure comparable to the conditions at the Earth's inner solid core. The latter simulation serves as a critical test of the generality of the method and demonstrates the importance of the magnetic effects in Fe in the Earth's core despite its extreme temperature and pressure.

The use of computational simulation is by now so pervasive in society that it is no exaggeration to say that continued U.S. and international prosperity, security, and health depend in part on continued improvements in simulation capabilities. What if we could predict weather two weeks out, guide the design of new drugs for new viral diseases, or manage new manufacturing processes that cut production costs and times by an order of magnitude? What if we could predict collective human behavior, for example, response to an evacuation request during a natural disaster, or labor response to fiscal stimulus? (See also the companion CCC Quad Paper on Pandemic Informatics, which discusses features that would be essential to solving large-scale problems like preparation for, and response to, the inevitable next pandemic.) The past decade has brought remarkable advances in complementary areas: in sensors, which can now capture enormous amounts of data about the world, and in AI methods capable of learning to extract predictive patterns from those data. These advances may lead to a new era in computational simulation, in which sensors of many kinds are used to produce vast quantities of data, AI methods identify patterns in those data, and new AI-driven simulators combine machine-learned and mathematical rules to make accurate and actionable predictions. At the same time, there are new challenges -- computers in some important regards are no longer getting faster, and in some areas we are reaching the limits of mathematical understanding, or at least of our ability to translate mathematical understanding into efficient simulation. In this paper, we lay out some themes that we envision forming part of a cohesive, multi-disciplinary, and application-inspired research agenda on AI-driven simulators.

Direct observation of chemical reactions is notoriously difficult. Reaction rates tend to be too fast for chemists to be able to see how molecules move as they combine and change, and individual electrons -- the species that are directly involved with reactions-- are subject to the laws of quantum mechanics that make direct observation of their position impossible. It'd especially difficult to observe reactions between organic molecules involving catalysts, because the reactions can take place at extreme temperatures and pressure, often proceed via very short-lived and unstable intermediates formed by combinations of the reactants with the catalyst. This makes it difficult to determine the mechanism of the reaction, which in turn complicates the design of improved catalysts. An interdisciplinary team of chemists, physicists and computer scientists at the US Department of Energy's Brookhaven National Laboratory in New York State and nearby Stony Brook University have devised a method to analyse data from X-ray crystallography to decipher the three-dimensional nanostructures that form during catalysed reactions.

The longest amount of time that a spacecraft has survived on the surface of Venus is 127 minutes. On March 1, 1982, the USSR's Venera 13 probe parachuted to a gentle landing and managed to keep operating for just over two hours by hiding all of its computers inside of a hermetically sealed titanium pressure vessel that was pre-cooled in orbit. The surface temperature on Venus averages 464 C (867 F), which is hotter than the surface of Mercury (the closest planet to the sun), and hot enough that conventional electronics simply will not work. It's not just the temperature that makes Venus a particularly nasty place for computers--the pressure at the surface is around 90 atmospheres, equivalent to the pressure 3,000 feet down in Earth's ocean. And while you can be relieved that the sulfuric acid rain that you'll find in Venus' upper atmosphere doesn't reach the surface, it's also so dark down there (equivalent to a heavily overcast day here on Earth) that solar power is horrendously inefficient.

The longest amount of time that a spacecraft has survived on the surface of Venus is 127 minutes. On March 1, 1982, the USSR's Venera 13 probe parachuted to a gentle landing and managed to keep operating for just over two hours by hiding all of its computers inside of a hermetically sealed titanium pressure vessel that was pre-cooled in orbit. The surface temperature on Venus averages 464 C (867 F), which is hotter than the surface of Mercury (the closest planet to the sun), and hot enough that conventional electronics simply will not work. It's not just the temperature that makes Venus a particularly nasty place for computers--the pressure at the surface is around 90 atmospheres, equivalent to the pressure 3,000 feet down in Earth's ocean. And while you can be relieved that the sulfuric acid rain that you'll find in Venus' upper atmosphere doesn't reach the surface, it's also so dark down there (equivalent to a heavily overcast day here on Earth) that solar power is horrendously inefficient.