University of Virginia Main Campus

NSF Awards · FY 2025 · 2025-09

This project focuses on designing new methods to facilitate the evaluation of artificial intelligence (AI) agents. It the era where AI agents are rapidly proliferating, with new systems of increasingly capable AI technologies, it is crucial to thoroughly understand their performance capabilities and limitations—a prerequisite for both safe deployment and continuous improvement. Traditional evaluation methods require running AI agents in live environments to collect performance data, but this approach can be resource-intensive and pose significant safety risks. This project addresses these challenges by developing innovative evaluation methods that dramatically reduce the need for expensive and potentially hazardous live testing, thereby accelerating the safe deployment of current AI systems and enabling the development of next-generation AI agents. Additionally, the project will train future AI researchers, helping to expand access to AI research opportunities across the United States. This project pioneers three research thrusts to fulfill different evaluation needs. First, the project delivers methods to efficiently evaluate an AI agent in a holistic manner with a scalar performance metric by reimagining Monte Carlo methods. The key innovation involves repurposing offline data to inform the online sampling process of Monte Carlo methods, thereby reducing the required sample size for accurate performance estimation. Second, the project develops methods to efficiently evaluate an AI agent in a fine-grained manner across different initial conditions by reinvigorating value function learning. The approach identifies statistical metrics that are most indicative and influential to the performance of the learned value function, then optimizes those metrics during online data collection to maximize evaluation efficiency. Third, the project delivers methods to efficiently evaluate an AI agent according to human feedback by developing transformative techniques that substantially improve reward model quality while minimizing human annotation requirements. Through these research activities, the project aims to significantly enhance current methodologies for evaluating AI agents, ultimately accelerating the development pipeline of AI systems. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

This NSF award, “A photonic quantum simulator of quantum field theory and spin physics,” will tackle the co-design of specialized quantum computing processors, termed quantum simulators, to solve specific, hard physics problems such as quantum field theory in subatomic physics and quantum magnetism in condensed-matter physics. Quantum computing is set in an ultracompetitive international context, promising deeply disruptive exponential speed-ups of intractable classical tasks such as quantum simulation (whose applications such to nitrogen fixation or superconductivity have societal importance) as well as code-breaking by integer factoring (which directly impacts national security). While quantum computing is now the object of large-scale industrial efforts focused toward commercial applications, academic “blue sky research” remains crucial to explore more easily reachable, specialized, yet powerful quantum machines in the service of the advancement of fundamental science. The co-design effort will consist of conceptualizing quantum photonic circuits, specialized in the implementation of quantum evolution per specific Hamiltonians of interest, as opposed to generic universal quantum gates. This approach is based on recent advances by the PI’s group, generating record-size entanglement in cluster quantum states which constitute complete substrates for measurement-based quantum computing. This large-scale entanglement directly translates into the quantum computing “volume” (number of qubits times circuit depth). The other essential component is the generation of nonlinear (“non-Gaussian”) quantum gates to simulate quantum evolution that is hard to calculate classically, such as the seminal quartic phase gate that describes the self-interaction of the Higgs field in the standard electroweak theory. The PI’s research has shown that a central resource to enable the realization of non-Gaussian gates is photon-number-resolving (PNR) detection, a sophisticated technical resource featured in the PI’s laboratory. Recent results by the PI’s group show that cluster states and PNR detection can be used to form specific quantum optical circuits, driven by a neural network that was trained by reinforcement learning. Such circuits are predicted to generate the desired non-Gaussian quantum gates with success probability rates above 95%, far outperforming all other proposals. This fundamental research on co-designed quantum photonic circuits will pave the way to subsequent implementations at very large scale in integrated “on-chip” photonics. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

In the “Beyond Moore’s Law” era with increasing edge intelligence, domain-specific computers in heterogeneous fabrics will rule the roost. Algorithms accelerating NP-hard (i.e., provably complex) applications or pre-compute processes that do not demand exact precision will run on tailored hardware. The hardware performance, rather than the algorithmic or software efficiency, may dictate solution speed, energy cost, footprint, and cyber-resilience. Clever hardware innovations for application-specific integrated circuits (ASICs) are no longer a rarity, but they all employ conventional material platforms like silicon, insulators, and compound semiconductors. This proposal will explore a new prospect – the use of quantum materials with exotic properties – to elicit computational activity with unprecedented size, weight, and power (SWaP). Additionally, innovative technologies and methods to train students in lab procedures through virtual platforms (e.g. GoPro video sessions, kid-friendly Minecraft and Roblox design challenges) will be developed and posted on YouTube and Vimeo for the public. Students selected through online exercises will be evaluated using rubrics developed by learning centers at the universities and sent to the Army Research Laboratory (ARL) and the National Institute of Standards and Technology (NIST). For the hardware needs of modern computing and artificial intelligence to be “self-contained”, all the data and resources needed to execute a computing task should be available in situ and not have to be fetched from a remote server or “cloud” which may be unreliable or unavailable. One powerful paradigm that satisfies many or all of these requirements is “processor-in-memory (PiM)”, where compute happens right at the memory site. The project plans to design, simulate, fabricate, characterize, and experimentally demonstrate a processor-in-memory architecture implemented by heterogeneously integrating a topological insulator (TI) (a quantum material) with nanomagnets and a piezoelectric material. The nanomagnet enables storage and the piezoelectric enables gating, while the TI brings in both high spin-selectivity and voltage tunable bandgap. The device is projected to perform logic operations and image processing with ultralow footprint and energy cost. A new and powerful PiM-based on a novel genre of materials with unusual quantum mechanical properties will be developed, which can be leveraged to outperform other PiMs in energy consumption, footprint and speed. This PiM will be built, characterized and its superior performance demonstrated. New light will be shed on the physical properties of these quantum materials to stimulate further research to benefit computing hardware. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

Massive stars are born in thick clouds of gas and dust. Within these environments, radio telescopes have revealed compounds known as “complex organic molecules” (COMs). These carbon-based molecules are precursors of larger molecules that are needed for life. Understanding how COMs form and evolve in space is key to uncovering the origin of complex chemical species on Earth and on planets orbiting other stars. Led by a team at the University of Virginia, this project combines cutting-edge astronomical observations and computational chemical simulations to investigate how and why the chemical content varies across different star-forming environments. This research will help astronomers to understand the chemistry of our galaxy and to trace the building blocks of life across the universe. In addition to expanding our knowledge of space chemistry, the project also focuses on training future scientists. It will support students at every level, from elementary school through graduate school, and increase learning opportunities for students with limited exposure to astronomy and space science. Public events will help bring science education and excitement to local communities. Understanding the origin and evolution of space chemistry, particularly the disparities in chemical content observed between star-forming regions, requires a comprehensive study of a large sample of star-forming objects spanning a wide range of masses, ages, and environments. This project combines (i) multi-wavelength observations of molecules in massive star-forming regions with (ii) state-of-the-art simulations that provide a theoretical picture of the spatial distribution of molecules in such regions at different moments. From here, the team will build a detailed chemical evolutionary sequence for massive star formation, guided by observations that resolve molecular emission across various sources at different scales and evolutionary stages. Synthetic spectra and emission maps generated from the simulations will be directly compared with observational data to identify the chemical pathways leading to various important molecules and to determine how environmental factors shape the chemical content of new stars and planets. This project will directly inform the interpretation of data from radio-telescopes such as the NSF’s ALMA and Green Bank Observatory (GBO). This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

The analysis of imaging outcomes is a dynamic and rapidly evolving research field, driven by the growing accessibility of large-scale biomedical imaging databases. Imaging data, often characterized as functional data, presents unique opportunities and challenges for statistical analysis. Existing methods, however, are insufficient for handling the computational demands of large-scale medical imaging data or addressing issues such as unmeasured confounding and population heterogeneity in causal analysis. This research will develop advanced statistical tools to overcome these critical hurdles. By developing new techniques that efficiently process large-scale imaging information and provide more accurate causal insights, this work will advance national interests in scientific innovation and evidence-based decision-making. It will promote scientific progress in a vital area of imaging data analysis and aims to advance public health by enabling a deeper understanding of treatment effects from observational studies. The developed data analytics tools also have broad applicability across various fields, including aging research, digital health, and plant science, addressing challenges faced by modern society. Furthermore, the project will benefit the broader research community through the release of freely available software tools and will support STEM education by involving undergraduate and graduate students in hands-on research and integrating project findings into curriculum development. This project aims to develop a general functional data analysis (FDA) framework for analyzing large-scale imaging data and uncovering causal relationships between treatments/exposures and imaging responses. Specifically, the project will address challenges in large-scale observational imaging studies via three aims. First, it will develop functional regression models for imaging responses based on a distributed learning framework, enabling scalable yet accurate estimation and inference. Second, it will introduce an image-on-scalar instrumental variable regression to mitigate confounding bias in observational studies. Third, it will propose an image-on-scalar doubly robust regression method leveraging functional pseudo-outcomes to address population heterogeneity. The proposed methods will be rigorously evaluated using existing imaging studies and are expected to significantly advance the methodology, theory, and computation of FDA and causal inference. Additionally, by releasing open-source software, the project will empower researchers to harness vast amounts of imaging and functional data from publicly available repositories. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

Infectious diseases are widespread and reduce the health of their hosts. Hosts thus experience strong selection to avoid contracting an infection. This research will advance our understanding of the evolution of host strategies to avoid contact with parasites. Researchers for this project propose that hosts avoid their parasites through dispersal, the permanent movement between sites. This idea is important because it asserts that dispersal, a widespread behavior, drives rates of contact with parasites and evolves as part of hosts’ defense against infection. The proposed work will use experiments and field surveys to test parasites as drivers of dispersal evolution. The project will also create educational programs that engage community college transfer students in scientific research. The main idea of this work is that dispersal strategies evolve to be sensitive to infection risk. The researchers predict that hosts evolve to disperse early in an epidemic, when infection risk is low. The work will conduct experiments with a free-living nematode and its natural parasites. The researchers will track the evolution of dispersal strategies under parasite selection and identify the genetics and mechanisms of adaptation. They will also test if dispersal limits selection for other defenses. Finally, they will assess whether dispersal reduces infection risk in the wild, by observing the natural distribution of parasites. Community college students will contribute to this work through a research course at the local community college and a summer research experience for incoming transfer students. A bridge course will connect transfer students with research and provide academic support. Community college transfer students are more likely to identify as low-income and first-generation, and this work will promote their success in STEM while advancing fundamental knowledge of parasites as selective agents. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

The aim of this project is to analyze how centers influence scientific collaboration across U.S. research universities, identifying the structural and organizational factors that predict their success. University research centers and institutes play a pivotal role in advancing interdisciplinary collaboration and knowledge production. These entities are often designed to break down disciplinary boundaries and promote innovation. This research fills a gap in knowledge of their long-term impact on researcher networks, institutional structures, and scientific outcomes. The project serves the national interest by improving the effectiveness of research investments—guiding universities, funders, and decision makers in designing more effective research environments—and by developing research methods using artificial intelligence. Public-facing tools, including an open database, make these insights broadly accessible to decision-makers. This work also strengthens U.S. research infrastructure by equipping institutions with evidence-based strategies to support interdisciplinary science, translational research, and the next generation of cutting-edge research centers. This study builds a comprehensive, longitudinal dataset of research centers and institutes at over 300 U.S. universities linked to affiliated faculty, funding sources, and collaborative research outputs. Using this dataset, the project (1) maps center affiliations and associated collaboration networks, (2) classifies centers by their founding collaboration structures and analyzes how those relate to long-term performance, interdisciplinarity, and scientific impact, and (3) uses natural experiments to estimate the causal effects of center affiliation on individual collaboration practices and interdisciplinary research production. These aims are accomplished through advanced computational and AI techniques, including affiliation disambiguation algorithms, network analysis, large language models, and quasi-experimental methods. The resulting empirical and methodological contributions inform theories of collaboration, organizational science, and the science of science, while also generating practical tools and guidance to support the future of interdisciplinary research. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

This award supports a program of experimental research by the University of Virginia group, to explore the limits of our knowledge of certain subatomic particles, and of their interactions at the smallest length scale. The project investigates three systems: the pi meson, the muon (a more massive "cousin" of the electron), and the neutron. The research program is focused on the intrinsic low-energy properties of each of them, properties that are sensitive to some of the basic tenets of the Standard Model (SM). The SM describes all known (to date) elementary particles, their basic characteristics, and their interactions. Determining these properties precisely enough to establish whether they differ from SM predictions has the potential to uncover new physical processes and particles. The project aims to advance basic science in nuclear and elementary particle physics, contribute to building the nation's STEM talent by educating members of the new generation of the STEM workforce, while tapping the broadest available talent pool. Specifically, in the course of the award period, it is anticipated that two graduate students will complete the requirements for a Ph.D. degree in physics, and another one will graduate with an M.Sc. degree. Finally, the project aims to develop new tools in fundamental research with possible broader applications. The proposed research continues an ongoing program of study of the electroweak and strong interactions at low energies. The immediate goals of the project are: (a) carrying out the measurements of the Nab experiment, a program of precise measurements of the neutron beta decay parameters at the Spallation Neutron Source (SNS), Oak Ridge National Laboratory; (b) finalizing data analysis and publication of results on the muon gyromagnetic ratio determined in the Muon g-2 experiment at Fermilab, (c) analyzing the Phase-I results of the MUonE experiment at CERN, and (d) conclusion of the data analysis and publication of results for the PEN experiment, a precise measurement of the electronic and radiative electronic rare decays of the pion. Nab is a high-rate unpolarized measurement of the electron-antineutrino correlation parameter "a" with accuracy of a few parts in 1000, and a precise measurement of the neutron decay parameter "b", the Fierz interference term. Nab aims to resolve the persistent inconsistencies in the determination of the nucleon axial vector form factor and, consequently, of the Vud term in the Cabibbo-Kobayashi-Maskawa quark mixing matrix. This will ultimately provide new limits on possible extensions of the Standard Model, such as left-right symmetric models, leptoquarks, etc., by exploiting the unique advantages of neutron decay, one of the most basic and theoretically best understood processes in nuclear physics. The Muon g-2 and MUonE experiments are focused on the hadronic corrections to the extremely precisely predicted electroweak processes that determine the interaction of charged leptons with the vacuum, a research subject with one of the highest potentials for discovery of beyond-SM physics. The Muon g-2 experiment has completed data taking and the main data analysis, and is in the process of publishing details of the analysis. The MUonE experiment at CERN, having established full viability, is wrapping up Phase-1 data acquisition. The goal of PEN is to reach the precision of 5 parts in 10,000 for the branching ratio of the electronic decay of the pion, which providess the most accurate experimental test of lepton universality (LU) available. At present, precision of the pi-e2 decay measurements lags behind the SM theoretical predictions by more than an order of magnitude. A number of physics scenarios outside the SM would lead to LU violation. LU, and lepton properties in general, carry added significance due to developments in neutrino physics This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

Near the end of their lives, Sun-like stars become luminous red giants – they expand over a hundred times in radius and their outer layers cool. These red giants pulsate, changing in brightness periodically, and they lose mass through thick, dusty winds. Planets and brown dwarfs (objects bigger than planets but smaller than stars) orbiting nearby can be strongly affected by these dramatic changes. This project will use supercomputer simulations to explore what happens to close-in planets and brown dwarfs as their host stars evolve. Can these companions survive in such extreme red giant environments? If so, how do their properties, such as mass, temperature, and composition, change over time? The project will also examine how planets and brown dwarfs, in turn, affect the red giants and their dusty outflows. In carrying out the research, PhD and undergraduate students will be mentored and trained not only in astrophysics, but also in high performance computing, data science, analysis and visualization. The research team will also build connections with the local community, creating opportunities to enhance public understanding of science. This project will advance understanding of how planets and brown dwarfs may influence, and be influenced by, their evolving stellar hosts. A key question to be addressed is whether the presence of planets and brown dwarfs can explain the mysterious "Long-Secondary Periods" (LSPs). LSPs are periodic dimming events that may be caused by low-mass companions with dusty, comet-like tails that block the stars’ light as they orbit. LSPs have been seen in tens of thousands of red giants. If planets and brown dwarfs are indeed responsible for these brightness changes, it would open a new window into a vast population of distant planetary systems in the late stages of evolution —systems that would otherwise be difficult to detect and study. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-08

Symmetry is a fundamental property of physical and mathematical objects. This project will investigate two forms of symmetry in mathematics: reflectional symmetry (e.g., flipping a circle across a diameter) and rotational symmetry (e.g., rotating a circle around its center). The PI will use modern techniques from algebraic topology to classify the rotational symmetries of certain high-dimensional geometric objects and understand their behavior under reflectional symmetries. In another direction, the PI will study how certain algebraic and geometric objects with reflectional symmetry appear in different areas of mathematics. The PI will also develop and study the effects of community-engaged pedagogy in undergraduate math courses, with an emphasis on K-12 education and outreach, continue co-organizing the Electronic Computational Homotopy Theory Online Research Community, and organize two regional workshops and a regional conference in the Mid-Atlantic. Specific research projects include the study of the stable homotopy groups of spheres using topological modular forms and applications of these results to problems in geometric topology and Riemannian geometry; namely, the classification of exotic spheres, the detection of their rotational symmetries, and the study of lower curvature bounds on their Ricci curvature. In another direction, the PI will continue developing the theory of homological stability for sequences of topological spaces equipped with group actions, with the aim of developing a general framework for passing results from the non-equivariant to the equivariant setting. Finally, the PI will continue investigating real algebraic K-theory, the K-theory of rings and ring spectra equipped with anti-involutions, using trace methods, with applications in stable homotopy theory and arithmetic geometry. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-08

With the support of the Chemical Catalysis program in the Division of Chemistry, Professor Jason Bates of the University of Virginia and Professor Brandon Bukowski of Johns Hopkins University are studying new approaches to improve the productivity and durability of molecular catalysts through supporting them on nanostructured solids. These fundamental approaches will enable more efficient, flexible, and sustainable pharmaceutical manufacturing processes by developing catalysts that are not only more active, but also longer lasting and safer to use in continuous flow systems. These advances will help facilitate a transition to a distributed manufacturing model that can respond quickly to shifting or localized demand. Combined experimental and computational studies in this project will focus on asymmetric hydrogenation, which is an important catalytic reaction in pharmaceutical production, and computational methods and models will be made broadly accessible to the community. Educational and outreach efforts will train undergraduate and graduate students in cutting-edge experimental and machine learning techniques and engage high school students through hands-on science programs. The team will build strong inter-institutional collaborations through regular student exchanges, helping to prepare the next generation of scientists and engineers. With the support of the Chemical Catalysis program in the Division of Chemistry, Professor Jason Bates of the University of Virginia and Professor Brandon Bukowski of Johns Hopkins University are studying the impact of solid support structure on heterogenized asymmetric hydrogenation catalysts. By integrating experimental and theoretical approaches, the project will establish a materials-driven approach to molecular catalyst design, leveraging the structural features of zeolites as tunable, non-coordinating solid supports for cationic complexes. Hierarchical and nanosheet zeolite structures will be targeted to exploit partial confinement at zeolite nanopore mouths while maintaining substrate accessibility. The team will combine synthesis, spectroscopic characterization, kinetic analysis, and ab initio microkinetic modeling to systematically explore how support structure influences catalytic behavior. These insights will inform the development of predictive computational tools to guide the design of zeolite–ligand pairs across a broad chemical space, to reshape reaction energy landscapes and suppress off-cycle deactivation pathways. Integrating advances from zeolite synthesis, molecular catalysis, and computational catalysis, this project offers a blueprint for discovering supported catalysts that outperform their homogeneous analogues. The experimental and computational approaches will be extensible beyond asymmetric hydrogenations, offering the potential to improve the reactivity of other key reactions facilitated by molecular catalysts, such as cross-couplings and selective oxidations. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-08

Due to the large volume of datasets and the stringent communication requirements by modern applications, the exchange of data for learning and computing purposes needs to be done in a timely manner. This project introduces the notion of age of information (AoI), used to assess timeliness in networks, into the study of federated learning (FL), with the aim of providing low-latency and communication-efficient means for data exchange in large-scale FL systems. The proposal focuses on designing novel client scheduling, information quantization and client-server association methods to enable timely FL over wireless communication networks. This research is expected to result in significant broader impacts rendering large-scale deployment of real-time monitoring and information sharing systems using FL. It can potentially impact various applications, including collaborative autonomous driving, precision healthcare, and others. The algorithms, analysis, and experimentation developed will advance the state of the art in communication theory, networking, and machine learning, and would naturally translate into undergraduate and graduate courses taught by the PIs in these areas. The goal of this project is to design and analyze efficient agent scheduling policies and communication schemes that realize the notion of timely FL over communication networks imposing various system level constraints. It includes three principal thrusts. The first thrust focuses on developing various timely and low-latency agent scheduling policies, inspired by the AoI metric, and analyzing their convergence performances. To further improve the communication efficiency, the second thrust investigates novel joint model compression and scheduling approaches to enhance the communication efficiency over unreliable networks while maintaining reasonable FL performance. To cope with the dynamically evolving communication environment, the third thrust develops online learning based agent grouping and model aggregation approaches to enable timely hierarchical FL, where multiple servers are connected together through a hierarchical multihop network. Finally, a thorough validation of the developed algorithms will be performed using real-world datasets and a lab testbed. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-08

Dormancy is a survival strategy used by cells across all forms of life. When nutrients such as food and water are scarce, cells enter a low-energy state to preserve vital functions. This project will investigate how cells reorganize their machinery during dormancy, focusing on ribosomes which are molecular machines that build proteins. The research team, led by Dr. Ahmad Jomaa at the University of Virginia and Dr. Simone Mattei at the EMBL Imaging Center in Germany, recently discovered that in dormant yeast cells, ribosomes stop making proteins and hibernate by attaching themselves to fragmented mitochondria. This unexpected behavior suggests a new way that cells conserve energy during stress. The Jomaa lab will use advanced imaging and molecular biology techniques to uncover how ribosomes become inactive on mitochondria and what role this plays in helping cells survive under stress. These findings will provide fundamental insights into how life adapts to extreme conditions, with broader implications for medicine, agriculture, and biotechnology. The project also includes an educational outreach effort that researchers established called “Molecular Touch.” This program will introduce middle and high school students in small classes at the Charlottesville and Albemarle County schools to the world of structural biology using 3D-printed models of ribosomes. This initiative supports NSF’s mission by advancing discovery while expanding STEM education and workforce development. This project will investigate the molecular and cellular basis of dormancy, a reversible quiescent state that enables cells to survive under nutrient-limiting conditions. Dormancy is a conserved survival strategy observed across all domains of life, yet the mechanisms that regulate energy conservation and repression of protein synthesis during this state remain poorly understood. Recent work by the Jomaa lab at the University of Virginia and the Mattei lab at EMBL revealed a novel phenomenon in dormant Schizosaccharomyces pombe cells: Ribosomes, which normally produce proteins in the cytoplasm, become stably tethered to mitochondria and enter an inactive, hibernating state. This discovery suggests a spatially organized and regulated mechanism of translational repression during dormancy. The two teams will combine cryo-electron tomography, yeast genetics, biochemistry, and high-resolution structural biology to uncover the molecular determinants of ribosome-mitochondria tethering and its functional consequences. The specific aims are to: (1) Identify the mitochondrial receptor responsible for ribosome tethering during glucose depletion; (2) Determine how the ribosomal protein RACK1 regulates tethering, mitochondrial function, and stress survival, and; (3) Elucidate how ribosomes are released, and translation resumes upon nutrient repletion. By uncovering how cells reorganize the translational machinery during metabolic stress, this project will advance our understanding of cellular adaptation, protein synthesis regulation, and the molecular logic underlying dormancy and recovery. This project is funded by the Cellular Dynamics and Function program of the Molecular and Cellular Biosciences Division in the Biological Sciences Directorate. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-08

Nontechnical Description Two-dimensional (2D) materials are atomically thin sheets with atoms strongly bonded within each layer but weakly bonded between layers. This provides unique versatility in their electronic, quantum and topological properties that can be achieved by twisting of the sheets between layers or combining different 2D sheets to form heterostructures. They have shown promise for emergent technologies such as artificial intelligence, quantum computing and photonics. This project focuses on transition metal dichalcogenides (TMDs), a class of 2D materials with layers of transition metal such as molybdenum or tungsten terminated with a chalcogen (sulfur, selenium, or tellurium) on either side. Defect engineering and doping are critical to modulate their electronic and photonic properties but challenging to achieve due to the TMD structure and bonding. The goal of this project is to establish a new and universal approach to achieving precise control of TMD properties with electronic and magnetically active dopants. Success will realize novel magnetic and electronic 2D materials, and advance 2D device technology. To this end, investigators will develop a doping method termed “backdoor doping” where low energy ions are injected into 2D materials. This ideally embeds the dopant atom in a substitutional site while avoiding damage to the material. Developing a highly competent and technologically knowledgeable workforce is at the core of a university education. The technical work will be integrated with education and training of undergraduate and graduate students. Research projects will enable students to develop critical thinking, collaborative work ethics, and a unique set of skills. A team of undergraduate students will be integrated in the EAGER research and tasked with projects including problems related to data analysis, simulation of ion-matter interactions, and image analysis. Technical Description A universal method to dope TMDs and engineer defects is uniquely important and challenging at the same time. This EAGER project will validate the feasibility of “backdoor doping” where the projectile is a recoil target atom generated by irradiation of a thin metal target, ejected from the backside of the foil with low kinetic energy and then implanted in the TMD, as a means to achieve controlled doping with various transition metals, and to predict feasibility for a wide range of dopant-material combinations. Judicious choice of TMDs (light metal, heavy chalcogen or vice versa), varied projectile mass, and use of magnetic dopants is combined with modulation of the ion energies to achieve selective displacement. This leads to preferential replacement on the chalcogen or metal sub-lattices with well-defined bonding states characterized by their ligand field splitting. The challenges lie in generating the low energy ions for a wide range of elements, and to find the right conditions to achieve displacement without incurring unwanted damage. A combined experimental and computational approach will accordingly be used to understand the defect inventory generated by ion impact events. Atomic resolution scanning transmission electron microscopy (STEM) and scanning tunneling microscopy and spectroscopy (STM/STS) are instrumental to this study. The work will also advance the fundamental, mechanistic understanding of ion-TMDC interactions. The proposed “backdoor doping” removes the need for a large ion source, offers clean implantation environment limiting contaminant implantation, and provides high versatility in dopant-material combinations. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-08

This award provides funding for U.S.-based individuals to participate in the conference Young Mathematicians in C*-Algebras (YMC*A), to be held July 21 - 25, 2025 at Southern Denmark University in Odense, Denmark. This meeting is organized for and by graduate students and postdoctoral researchers in operator algebras and related areas, with the goal of fostering scientific interaction among early-career researchers. The conference also includes a panel focused on advice to help early-career researchers navigate the career path of mathematics research. In each of its previous seven editions, YMC*A has provided an excellent opportunity for over one hundred early-career operator algebraists from around the world to attend mini-courses on current research topics in operator algebras. This grant significantly boosts the participation of U.S.-based researchers and their institutions at this conference, exemplifying and enhancing U.S. research and furnishing opportunities for researchers to expand their professional networks. More information about the conference is available at: https://sites.google.com/view/ymcstara-2025. The conference focuses on recent developments in operator algebras, noncommutative geometry, and related areas of mathematical analysis. The conference features three mini-courses by established researchers alongside many contributed research talks by participants, and mentoring activities. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-08

This project aims to serve the national interest by organizing coordinated multi-site replication studies on open computing education research questions. Research relies on the ability to replicate studies to understand the limits and applicability of research findings and to build confidence in results. Replication studies are especially important in computing education research where results are drawn from human subjects studying computing. The goal of this IUSE:EDU level two Institutional and Community Transformation project is to create cohorts of researchers to design replication packages for rigorous research that will support cross-study meta-analysis, build fundamental knowledge on key topics, and provide the foundation for computing education research theory building. The project will involve approximately ten core replication designers and forty replication participants impacting thousands of students across the United States. A replication design team will be convened to create replication packages that address research goals and questions of interest to the computing education research community. A project team will then be formed to recruit a series of replication participants who will run the study in their context using the replication packages produced by the replication design team. Members of the computing education research community will be invited to run the replications in their classrooms, contributing to the overall investigation into the topic. Results of these studies will address common research questions and lay the foundation for computing-specific educational theory. The results of the coordinated, multi-site replication studies will provide deeper insights into important computing education research topics and will provide a better understanding of the impacts of various contextual factors on improving student learning and success in computing. The project evaluation will consist of formative and summative assessments of the replication process and objective measures such as publications and citations. The NSF IUSE: EDU Program supports research and development projects to improve the effectiveness of STEM education for all students. Through the Institutional and Community Transformation track, the program supports efforts to transform and improve STEM education across institutions of higher education and disciplinary communities.  This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-08

Large language models (LLMs) have transformed how people interact with artificial intelligence (AI) systems, achieving state-of-the-art results in various tasks such as education and healthcare. Their ability to understand and generate human-like text has opened up new possibilities for advancing scientific research, and one of the most promising applications of LLMs in this context is research idea generation, where LLMs can be utilized to identify novel research directions by analyzing existing knowledge. Although LLMs may generate claims that are not fully supported by existing publications, it is possible for them to generate potentially innovative research ideas that might have been overlooked by scientists working alone. Although quite a few studies have yielded promising results, significant advances are required for LLMs to generate accurate, easily verifiable, and trustworthy responses for scientific research. Moreover, for foundation models to achieve a comparable impact in research ideation, it is crucial that they are capable of optimizing both external and internal knowledge sources. Addressing this need, this project develops approaches that aim to effectively optimize two types of knowledge: external knowledge, drawn from diverse data sources, and internal knowledge, the parametric understanding acquired during training. A dual framework solution is designed for this optimization using agentic AI reasoning techniques. This research can improve the adaptation of external and internal knowledge of foundation models and their utility for scientific tasks. To optimize knowledge utilization in foundation models to foster research ideation, this project conducts three innovative research tasks: (1) The project develops an adversary-based reasoning approach to effectively harnessing the vast parametric knowledge within LLMs to improve research ideation. The project introduces adversarial learning in inference time, a paradigm shift from prompt design that unlocks the full potential of parametric knowledge of LLMs without requiring additional training. (2) This project develops a reinforcement learning-based approach to optimize LLM’s idea generation capabilities by leveraging external knowledge to systematically improve the utilization of external knowledge for idea generation. The project formulates the interaction between language agents and the external knowledge bases as a nested Markov Decision Process (MDP), where the outer MDP governs high-level action generation through interactions with the information retrieval environment, while the inner MDP controls token generation within LLMs. (3) This project develops a knowledge-based hallucination detection framework that assesses the groundedness of the generated research ideas and identifies hallucinated claims by analyzing the rationale behind the idea generation. The project also designs metrics to assess whether the ideation approach improves the quality of research ideas generated in terms of novelty and feasibility, and conducts extensive experimental studies to evaluate how the ideation approaches work using a variety of existing LLMs to generate research ideas based on the given approaches and evaluation metrics. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-07

This Smart and Connected Health (SCH) award supports research into new wearable robotic systems to enhance people’s quality of life. Musculoskeletal disorders (MSDs) are injuries or pains affecting the human musculoskeletal system. MSDs are one of the largest occupational problems in the United States, which significantly impact people’s quality of life and ability to work. Among them, upper extremity musculoskeletal disorders (UEMSDs) are one of the most common work-related health issues due to the physical performance of work. This research looks to develop new garment-like wearable robots for UEMSD prevention and rehabilitation. The research products will benefit the broader society by providing effective and accessible wearable robots for personalized therapies as well as training the US STEM workforce. The goal of this project is to understand the biomechanics pathology of UEMSDs and develop a new class of wearable robots considering biological characteristics and biomechanics for UEMSD prevention and rehabilitation. This project articulates biomechanics modeling, engineering design, deep learning, and ergonomics toward a wearable robot aided occupational assistance and rehabilitation protocol via four research thrusts. Firstly, this project seeks to understand the biomechanics pathology of UEMSDs and develop computational musculoskeletal models with a focus on occupational tasks. Secondly, the project looks to develop new garment-like wearable robots, i.e., wearable intelligent soft exosuits, considering the biological characteristics and biomechanics pathology of UEMSDs. Wearable sensors look to also be integrated into the wearable exosuits to provide real-time monitoring and estimation. Thirdly, the project seeks to establish wearable exosuits enabled UEMSD prevention and rehabilitation protocol and provide short- and long-term estimations of regained functional outcomes of users. Lastly, the project looks to also study the usability of the proposed wearable exosuits to improve the user-centered design. This project will generate new knowledge in understanding the biomechanics pathology of UEMSDs and new wearable robot enabled therapies for this significant health issue. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-07

This project improves the sustainability of the smartphone ecosystem. Smartphones have fundamentally changed the modern world, but their manufacturing, use, and disposal have also caused grave concern about sustainability, thanks to the unprecedented scale of their usage today. The project’s novelty is devising new sustainability-aware software for the smartphone ecosystem. The project’s broader significance and importance are the extension of smartphone usage lifetimes, the mitigation of the environmental impact of aging smartphones, and the introduction of sustainable computing as a new topic into existing Computer Science curriculum at the SUNY at Binghamton and at the University of Virginia. This project comprises three research thrusts: (1) a new operating system principal that continuously measures, accounts for, and rewards the sustainability impact of software activities; (2) new runtime systems that identify, circumvent, and survive software activities that may cause erratic shutdowns with aging batteries; and (3) cross-stack system designs that repurpose legacy smartphones for application-specific uses, and assemble multiple broken smartphones with complementary hardware. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-07

Combinatorics is an area of mathematics concerned with analyzing, organizing, and optimizing over discrete data. It is a fundamental tool in many scientific areas such as genomics, computer science, statistics, and physics. This project will develop combinatorial methods for attacking problems in Lie theory and symmetric function theory, areas with applications to probability, statistical mechanics, and quantum information theory. A mutually beneficial component is the further development of the SAGE open-source mathematics software, a crucial tool for this investigation. Also graduate students will be trained as part of this project, This project addresses combinatorial problems tied to representation theory, algebraic geometry, and physics, with a focus on Macdonald polynomials and Schubert calculus. Macdonald polynomials are a remarkable family of orthogonal polynomials which form a basis for the ring of symmetric functions. Since their introduction in the 80's, they have developed connections with many areas, including Hilbert schemes, the Calogero-Sutherland model in particle physics, and knot invariants. In the 90's, Garsia and Haiman studied transformed versions of Macdonald polynomials, which they connected to the representation theory of polynomial rings, generalizing classical results of Chevalley, Shephard-Todd, and Steinberg on reflection groups. A separate line of work initiated by Cherednik, Macdonald, and Opdam in the 90's investigated nonsymmetric versions of Macdonald polynomials, which clarified the theory and connected it to affine Hecke algebras. The PIs and collaborators recently discovered a way of transforming these nonsymmetric versions in the same style as Garsia and Haiman did for the symmetric case. Further study of these new polynomials will unearth new representation theoretic and combinatorial mysteries of nonsymmetric Macdonald polynomials. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-07

With support from the Chemical Catalysis program in the Division of Chemistry, David Flaherty (Georgia Institute of Technology), David HIbbitts (University of Florida), and Ayman Karim (Virginia Polytechnical Institute) will examine the connections among the structure, dynamics, and catalysis of reactions with oxygen on bimetallic nanoparticles. The team will create, characterize, simulate, and test how atoms of distinct metals move and facilitate reactions upon the surfaces of nanoparticles comprised primarily of gold with small amounts (1-5%) of a second element such as palladium or platinum. These materials are commonly described as single atom alloy (SAA) catalysts. These materials offer high rates and selectivities for numerous reactions important for domestic production of energy carriers and platform chemicals (e.g., valorization of biomass, shale gas, operation of fuel cells and electrolyzers). SAA currently suffer from a distressingly low number of active sites per gram of precious metal used. The collaborative team aims to develop methods to create SAA nanoparticles with smaller diameters (< 2 nm) to remedy this problem, and then test if the emergent and beneficial catalytic properties of these SAA are preserved as the size of the nanoparticles decreases. Here, the team will combine cutting-edge methods in quantum chemical simulations and multiscale modeling, characterization of operating catalysts using synchrotron methods, and catalyst testing and spectroscopy to learn how the nanoparticles restructure in different combinations of reactive gases relevant for catalysis (e.g., oxygen, hydrogen, carbon monoxide). Subsequently, the team will assess how rates and selectivities for a testbed reaction (reduction of oxygen with hydrogen) depend on the spatial organization of the atoms on the nanoparticle surface. Methods that will be developed will be useful for other dynamic catalyst systems and will be integrated into graduate-level courses. The proposed work involves lab-based education of graduate and undergraduate students and focused efforts to increase participation of women in catalysis science, especially with NSF REU (Research Experiences for Undergraduates) opportunities and cross-training of researchers across the three partnering institutions. Under this award, the collaborative Flaherty/Hibbitts/Karim team aims to learn how the structure, dynamics, and catalytic properties of bimetallic and SAA materials depend upon mean particle diameters, composition, and support identity, all factors that impact the coordinative saturation of surface atoms and the identity of their nearest and next-nearest neighbor atoms. The team will couple precise synthesis, advanced characterization techniques (including n situ, operando X-ray absorption spectroscopy, microcalorimetry, infrared spectroscopy), and computational methods (simulations of full nanoparticles with density functional theory and kinetic Monte Carlo) to address the complexity and dynamics of SAA catalysts. A testbed reaction system with rates and selectivities proven to be structure-sensitive with respect to these materials (H2 + O2 → H2O2) will be used to probe the surface structures of active catalysts, a challenge as the high pressures and complex solvents used often render characterization difficult. First, Au-rich bimetallic alloy nanoparticles (i.e., M1Aux materials, where M = Pd, Pt, Rh) with mean diameters of 1-2, ~6 and ~10 nm will be created, their post-synthesis structures will be characterized, and then the influence of adsorption and reactions on their structures will be examined over extended periods. Second, the thermodynamic relationships among adsorption energies, active site motifs, and nanoparticle structure will be determined. Third, the fundamental connections surrounding elemental identity, mole fraction, and coordination of the reactive metal and reaction rates, selectivities and barriers for H2O2 and H2O formation will be examined. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-07

This project provides support for 25 United States (U.S.)-based graduate students to participate in the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2025), to be held in Toronto, Canada, during August 3-7, 2025. KDD is one of the premier conferences in data mining, data science, and knowledge discovery in the world. It is a prestigious interdisciplinary conference that brings together researchers and practitioners every year. KDD covers topics in the data mining lifecycle (including algorithms, software, systems, and applications), as well as artificial intelligence, machine learning, data management, and information retrieval. Students participating in the conference are exposed to the latest research developments and can attend hands-on workshops, tutorials, eye-opening keynotes, and presentations. A strong representation of U.S. students and researchers is essential for maintaining U.S. competitiveness in these important areas. This project provides support for 25 U.S.-based graduate students to participate in the 31st ACM SIGKDD Data Mining workshop. The selection committee will select recipients of the support based on merit in the selection process. As an interdisciplinary conference, KDD attracts researchers and practitioners from academia, industry, and governments who work on all aspects of data mining and data science problems. It includes a highly competitive technical program, with regular peer-reviewed papers in the form of oral and poster presentations, as well as panel discussions and invited talks by leading experts in academia and industry. The conference places special emphasis on supporting students through training and mentoring by offering both undergraduate and doctoral student consortiums. These will provide a comprehensive multi-facet learning experience for students at all levels. This grant aims to help U.S.-based students overcome financial barriers that may prevent them from attending the conference. The award will be advertised on different sites and social media platforms, and the results will be announced on the KDD 2025 website (https://www.kdd.org). This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-06

As robots become increasingly helpful in factories, hospitals, and homes, they must learn how to interact effectively with people. When people communicate, they use words and body language—like hand movements and eye contact—to share what they mean. In the same way, robots that work closely with people need to understand both verbal words and nonverbal signals. One major challenge in creating such robots is that people’s behavior can change over time. For example, a person may use different gestures (e.g., pointing vs eye gaze) to describe an object at different times. The robot must notice these changes and adjust its responses. Also, if the surroundings change, robots must change how they act to keep being helpful. This project will help robots understand people better by studying both verbal messages and nonverbal cues. It will also improve how robots learn and plan to adjust when people’s behavior changes. The results of this research will impact people and help them to be more accepting of robots. The project will help students in several ways. First, new topics will be added to robotics classes. Second, the research will involve high school and college students. Finally, the team will share robotic advancements with the public through outreach activities. This project will focus on three goals to address challenges in multimodal human-robot interactions. First, it will develop advanced algorithms to understand human intentions. These algorithms will use multimodal data, such as verbal messages, hand gestures, eye gaze, and head movements. Second, the project will develop new learning and planning methods for robots. These methods will help robots plan and act in ways that align with human intentions. Lastly, the research will design adaptive learning techniques. These techniques help robots adapt to changes in human behavior to support long-term interactions. The proposed system will be validated through several comprehensive human-robot interaction studies. This project will move the HRI field forward by showing how we can create effective and adaptable long-term interactions between humans and robots. The results of this research will impact how people perceive and accept robots that will be more prevalent in their lives. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-06

Digital computing has been the backbone of the modern information revolution, driven by sustained advancements in digital hardware. Despite this tremendous success, many computational problems relevant to rapidly advancing fields such as machine learning, biotechnology, and resource optimization continue to challenge these platforms in terms of performance and efficiency. With the ever-increasing demand for computation and the slowing down of Moore’s law, there is a strong impetus to explore alternate computing paradigms and hardware platforms to solve such problems more efficiently. Probabilistic bit (p-bit)-based computing platforms, which exhibit characteristics starkly different from the deterministic behavior of digital bits, offer a promising pathway in such scenarios. However, current p-bit-based computing platforms have focused on a narrow set of functionalities, thereby limiting their broader potential. The goal of this project is to develop PRSIM, a new probabilistic computing engine with novel functional capabilities that can transform the efficacy of probabilistic computing. These advancements will be enabled and supported by novel probabilistic hardware realized using oscillator networks. PRISM will facilitate fundamental advances in probabilistic computing that will help overcome the constraints of existing platforms. These advancements will also have transformative downstream benefits for practical applications that require solving such problems. Computing platforms based on p-bits provide a natural hardware solution for accelerating Monte Carlo algorithms, a powerful tool for solving computationally intractable problems efficiently. However, existing platforms employ p-bits which are designed to probabilistically switch only between two states. This limitation can hinder the functional and performance capabilities of such platforms, especially when addressing problems that require more than two states. Currently, solving such problems necessitates computationally expensive pre-processing techniques just to make the problem compatible with the underlying p-bit platform, which increases the effective problem size on hardware and may degrade the solution quality. This project proposes to develop and demonstrate PRISM, a new computing platform using p-bits that are capable of probabilistic switching between multiple states. To enable this, PRISM introduces a novel hardware implementation based on oscillators and their networks which are capable of supporting multi-state p-bits. The PRISM platform facilitates direct mapping of a broad class of computationally hard problems that entail more than two states. This enhances performance and efficiency by eliminating the traditional overhead associated with solving such problems using two-state p-bit engines. The PRISM platform will be developed through a cross-cutting effort that spans both hardware and algorithm. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-06

This project provides funding for students from U.S. institutions of higher learning to attend the 2025 IEEE International Conference on Smart Computing (SmartComp 2025). Smart computing is a multidisciplinary field focused on the design, development, and deployment of computing technologies that are context-aware, adaptive, and deeply integrated with human and environmental systems. Applications span a wide range of domains, including smart health, smart cities, smart energy, and smart transportation. Advancing innovation in these areas requires a highly trained and diverse workforce, critical to maintaining the United States’ long-term technological leadership and economic strength. SmartComp is a premier venue for cutting-edge research at the intersection of computing, sensing, and human-centered system design, aimed at improving the quality of life. Increased participation of U.S.-based undergraduate and graduate students at SmartComp 2025 will help cultivate a next-generation workforce equipped to address complex, real-world challenges through smart technologies. SmartComp 2025 will be held in Cork, Ireland, in June 2025. The conference features a robust technical program, including keynote talks, workshops, poster sessions, demonstrations, and extensive networking opportunities with leading researchers, providing an exceptional experiential learning opportunity for participants. This project will provide travel support for U.S.-based students who would otherwise be unable to attend, broadening participation and expanding educational and professional development opportunities for emerging scholars in smart computing. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.