Columbia University

NSF Awards · FY 2026 · 2026-10

Pre-trained AI models shared through open online repositories are becoming essential infrastructure for research, industry, and government. But this growing reliance also creates an important cybersecurity concern: just as traditional software can be attacked to include viruses or access backdoors, AI models can also be tampered with. This can lead to security breaches and errors in systems that rely on these pre-trained models. This project will develop methods and tools to help users verify whether a pre-trained AI model is trustworthy before it is incorporated into scientific workflows, operational systems, or other important computing environments. By improving the security of this emerging AI infrastructure, the project will help strengthen the U.S. research enterprise, support economic competitiveness, and improve the resilience of AI-enabled systems. The project will also advance education and workforce development by training students, providing research opportunities, and fostering collaboration among universities, industry, and other stakeholders. This project develops a novel approach to address three major security challenges in the machine learning (ML) model supply chain. The research integrates software engineering principles with machine learning techniques to systematically mitigate vulnerabilities during model selection, loading, and management. First, the team of researchers will tackle model spoofing, where adversaries upload malicious models using deceptive names. The project relies on novel anomaly detection schemes for naming conventions and architectural signatures to identify these threats. Second, the investigators will secure the model deserialization process. Because frameworks often use formats vulnerable to arbitrary code execution, the research will develop automated, least privilege deserialization mechanisms and define safe subsets for model loading runtimes. Third, the project will establish robust model lineage tracking to manage the risks of reusing models. The team will create a lineage graph data structure that combines static and dynamic analysis to trace model evolution and detect illicit modifications. By integrating these methods, the project provides a comprehensive defense system that enhances trust, integrity, and oversight in the open source model ecosystem. 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 2026 · 2026-09

Generative artificial intelligence can create realistic text, images, and other complex data, offering new ways to support science, education, industry, and public decision-making. However, these systems can also make errors that are difficult for users to detect or measure. As a result, organizations may not know when generated data can be trusted, whether a system works equally well across different populations, or how its performance changes over time. This project addresses these challenges by developing statistical tools that measure the reliability and uncertainty of generative artificial intelligence systems. These tools will help researchers, educators, businesses, and public institutions use synthetic data more safely and effectively. One important application is the creation of digital twins of student populations, which can provide safe testing environments for new educational technologies without exposing real students to untested systems. The project also broadens participation in science and technology through hands-on activities for K-12 students, new graduate courses, doctoral student mentoring, and community workshops. This project develops a statistical framework for evaluating and designing uncertainty-aware generative models. The investigator will pursue three connected research thrusts. First, the project will develop methods to measure the overall fidelity of black-box generative models using the concept of effective sample size, which provides an interpretable way to assess how much reliable information the synthetic data contain. Second, the project will extend these methods to provide local and context-aware uncertainty estimates for specific inputs, subpopulations, and changing environments. These estimates will help identify when a generative model is reliable and when its outputs require caution. Third, the project will develop statistically guided training strategies that improve generative models for counterfactual analysis and choice modeling. These strategies include artificial intelligence assisted articulation for modeling human needs and knowledge distillation methods for learning counterfactual relationships. By embedding statistical goals directly into generative modeling, the project will advance the use of synthetic data in settings where real data are scarce, costly, sensitive, or inaccessible. The expected outcomes include new theory, practical methodology, open source software, and educational resources that support the trustworthy deployment of generative models. 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 2026 · 2026-07

Non-technical Description Polymers are widely used but may have unintended negative consequences, e.g., they form microplastics (sizes between 1μm and 3 mm) and nanoplastics (sizes between 10nm and 1μm), collectively termed MNPL. It is well-known that MNPL formation is triggered by (rare) bond-breaking events caused by exposure to air, water/solvent or to UV radiation, or by small external forces (e.g., polymers stretched for packaging). However, a thorough understanding of how such Å-scale bond-breaking events lead to much larger-sized fragments remains elusive. Filling the critical knowledge gap of the factors affecting MNPL formation, which could yield optimal pathways to their mitigation, is the focus of the PIs work in this area. An essential tool in this study will be the generation of data on MPNL formation and its analysis using AI and machine learning to look for hidden connections between polymer structure and ambient exposure. Such information will help to improve advanced manufacturing using these kinds of polymers. The PI will educate K-12 students, and in particular their teachers, on MNPL, and their formation mechanisms. The PI has already taught a group of 17 K-12 NYC teachers) who themselves will teach other teachers. Continuing these interactions, the PI now proposes to develop educational modules, in collaboration with K-12 teachers, to illustrate how MNPLs are created by ubiquitous processes such as shaking water in a plastic bottle, flowing water through PVC pipes, or through tire wear. Technical Description Based on available evidence, it is postulated that polymer morphology, i.e., amorphous, semicrystalline, or rubbers, is a critical variable in the creation of MNPL under quiescent conditions. Ambient exposure, e.g., oxygen, water, help to weaken and remove the amorphous polymer (mortar) – the crystalline portions (bricks) are then freed from each other and can go into the surroundings as MNPL. The PI hypothesizes that tailoring the mechanically key stress carriers in the mortar phase through a multipronged experimental and machine learning (ML) approach will provide several, complementary means to understand and hence mitigate MNPL release. Specifically, it is postulated that the length and number density of these stress carriers, and their ability to re-form, are central to polymer failure and hence MNPL creation. It is thus posited that increasing the number of stress carriers, reducing their lengths and/or in situ crosslinking them can result in slower MNPL release. The second key focus area is on polymers under (constant) stress, which will speed-up polymer failure and hence MNPL release. Critically delineating these two topics is the central emphasis. The experimental results from this work will be input into a ML formalism, coupled to a SHAP analysis, to enunciate the critical variables for MNPL formation; in parallel, the optimal conditions to minimize them will be determined by using the genetic algorithm. The PI has a strong track record of creating pipelines for students into STEM futures, and the PI will continue such career development for junior researchers. Finally, a center on Polymers at the end of life has been created in collaboration with NYU and other NY city schools. 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 2026 · 2026-07

Electronic systems are increasingly required to operate in extreme environments such as the surface of Venus, inside nuclear reactors, geothermal wells, and hypersonic aerospace platforms. Conventional silicon-based electronics fail at elevated temperatures because their electrical properties degrade rapidly above approximately 125 °C. Gallium nitride, a wide-bandgap semiconductor, offers superior thermal stability, radiation tolerance, and high-speed performance, making it a promising material for next-generation electronics in harsh environments. However, while individual gallium nitride transistors have demonstrated short-duration survival at high temperatures, little is understood about how complete circuits, particularly analog and mixed-signal systems that combine amplification, timing, and signal processing, degrade during prolonged exposure to heat and radiation. This project seeks to establish the scientific foundation needed to design reliable gallium nitride-based circuits that can operate above 500 °C for extended durations. The research will enable compact, energy-efficient electronics for planetary exploration, advanced energy systems, and distributed sensing in extreme conditions. The project also integrates research with education by developing a new graduate course on harsh environment electronics, mentoring undergraduate and high school researchers, and expanding outreach programs that introduce students to semiconductor reliability and extreme-environment engineering. By combining mission-driven research with workforce development, the project advances both national competitiveness in semiconductor technology and equitable access to engineering education. The technical goal of this CAREER project is to develop a degradation-aware design framework for gallium nitride mixed-signal circuits operating under combined high-temperature and radiation stress. The research focuses on indium aluminum nitride/gallium nitride high electron mobility transistors and co-fabricated passive components including capacitors, resistors, and inductors. The project will couple device-level degradation physics with electrothermal modeling and compact model extraction to enable predictive circuit design. Manufacturing processes will be developed using customizable processes in the Columbia Nano Initiative cleanroom, enabling optimization of gate geometries, metallization stacks, and integrated passive structures for operation above 500 °C. Devices and circuits will be characterized from cryogenic temperatures to 1000 °C using in situ radio-frequency probing and long-duration furnace testing. Two benchmark circuits will be developed: a ring oscillator that serves as a digital timing element and embedded degradation monitor, and a multi-stage low-noise amplifier that enables evaluation of analog performance metrics such as gain and noise figure at elevated temperatures. These blocks will be co-integrated into a mixed-signal test platform to assess system-level performance under thermal cycling and neutron and electron irradiation at Columbia’s Radiological Research Accelerator Facility. Experimental results will be integrated with physics-based simulations, circuit-level compact modeling frameworks, and particle interaction modeling to establish quantitative design rules linking materials degradation mechanisms to circuit-level performance metrics for applications such as wireless communications. The outcomes will advance fundamental understanding of wide-bandgap semiconductor reliability and enable scalable design strategies for extreme-environment electronics. 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 2026 · 2026-06

As wildlife populations across the globe experience dramatic declines, the resulting shrinking population sizes often lead to inbreeding, which can severely threaten the health and long-term survival of a species by exposing harmful genetic mutations. Understanding exactly how mating between closely related individuals causes these negative health impacts, known as inbreeding depression, is a critical challenge for evolutionary biologists and conservationists striving to design effective wildlife management strategies. To address this fundamental issue, this project investigates a unique, extensively documented population of wild house mice that has been isolated in a natural barn environment for over twenty generations. Researchers will combine more than two decades of observational data on mouse behavior, physical development, and survival with advanced genetic sequencing of thousands of individuals to map exactly how inbreeding affects physical traits and social networks. Ultimately, by revealing the hidden genetic costs of inbreeding and exploring whether animals naturally alter their behavior to avoid it, this research will provide vital scientific justification and practical tools to guide wildlife conservation efforts, while also contributing broadly to our understanding of genetic diseases. This project advances NSF's priorities in Biotechnology. The core objective of this research is to comprehensively characterize the phenotypic, genetic, and behavioral consequences of inbreeding in a natural mammalian population. The project leverages an unprecedented longitudinal dataset from a free-living population of wild house mice that was founded by twelve wild-caught individuals in 2002. Over twenty generations, researchers have collected detailed phenotypic data, encompassing social behaviors, morphological measurements, survival rates, and reproductive outcomes, for more than 95% of the approximately 20,000 individuals in this closed population. To achieve the project goals, the research team has generated a highly detailed extended pedigree and genotyped over 2.2 million variants for 8,500 descendants. This was accomplished by sequencing the founder genomes at high coverage and imputing low-coverage sequence data from the descendants back to the founder haplotypes. Utilizing this robust genomic and demographic dataset, the project pursues three specific aims. First, the investigators will quantify the effects of inbreeding on reproductive output, survival, morphology, and temporal social network interactions, followed by genome-wide association studies utilizing a recessive model to precisely identify specific genetic loci contributing to inbreeding depression. Second, the project will identify strongly deleterious alleles that do not tolerate inbreeding by comparing observed homozygosity against pedigree-based simulations, detecting transmission distortion in parent-offspring trios, and validating candidate variants through controlled laboratory crosses. The functional nature of these heavily selected variants will be characterized and cross-referenced with human disease databases to identify shared genetic vulnerabilities. Third, the study will investigate behavioral inbreeding avoidance by developing temporal network models of social interactions and mating outcomes, comparing empirical inbreeding frequencies against simulated null models to determine if these mammals actively avoid mating with close kin. By integrating genomic, demographic, and behavioral data across multiple generations in a wild setting, this project will significantly advance the fields of evolutionary biology and conservation genetics. It will illuminate the specific genetic architecture underlying inbreeding depression, the efficacy of purifying selection in natural populations, and the role of social behavior in mediating genetic health, thereby providing broadly applicable models for managing and preserving endangered wildlife populations. 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 2026 · 2026-06

Volcanic arcs can produce large eruptions. However, there are many unknowns about how the magma at volcanic arcs is formed. In particular, it is unclear whether these silicic magmas form deep in the mantle or shallow in the crust. This study will analyze trace elements using cutting edge techniques to explore the conditions of melt formation. The results will be important for understanding the causes of catastrophic explosive arc eruptions. The project will support an undergraduate student and will produce outreach materials on volcanic eruptions for the general public. This study will integrate olivine trace element data, melt inclusion data, and Chromian-spinel inclusion data to explore conditions of melt formation in the mantle wedge. Cutting-edge techniques will integrate mapping with major and trace element analysis of magmatic olivines. The project will test two hypotheses: 1) Arc olivine trace elements allow for a deeper look into mantle wedge processes than possible by bulk rock compositions. 2) The Tonga-Lau olivine trace element patterns ground-truth hydrous partial peridotite melting. The study leverages existing samples and international collaborations. 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 2026 · 2026-05

Mosquitoes can spread serious diseases to people when they bite, including dengue, chikungunya, and malaria. As mosquito populations expand into new areas, more people are at risk. One effective strategy to reduce disease transmission is to reduce mosquito populations by stopping them from successfully mating and reproducing. Some programs already release sterile male mosquitoes to reduce population size. These males can mate with females, but no offspring are produced. However, scientists still do not fully understand how mosquito mating behavior works, which can limit how effective these strategies are. Our research aims to better understand what signals and behaviors control successful mating in mosquitoes. We are especially interested in how these behaviors may differ among mosquito groups living in different environments. To investigate this, we will collaborate with a Vector Control Agency in Florida to collect mosquitoes from different regions, where habitats vary and may influence how mosquitoes mate. This work will help improve mosquito control strategies by identifying when and where sterile male releases may be effective, especially in areas where female mosquitoes are unlikely to mate more than once. In addition, understanding the biology behind mating could reveal new ways to reduce mosquito populations, such as methods that “switch off” a female mosquito’s drive to mate. As part of this research experience, undergraduate trainees will assist with field collections and laboratory experiments, and will also participate in oral and written presentations, as well as mock interviews to prepare for graduate training as members of the STEM workforce. This project advances NSF’s priorities in Biotechnology. In Aedes aegypti, females typically mate once and store sufficient sperm to fertilize all eggs produced over their lifetime. During mating, males also transfer proteinaceous factors that induce long-lasting suppression of female receptivity, ensuring exclusive paternity. Although our group previously identified a regulator of short-term post-mating receptivity suppression, the mechanisms by which females detect mating status and initiate and maintain long-term refractoriness remain poorly understood. The goal of this project is to define the molecular and neural mechanisms that link mating status to sustained changes in female behavior. We hypothesize, based on preliminary data, that receptor-expressing cells in females detect male-derived proteins transferred during copulation and relay this information to neural circuits that suppress receptivity. To test this hypothesis, we will pursue three research aims: (1) identify receptors responsive to male-derived, paternity-enforcing factors using a multiplexed high-throughput in vitro screening platform; (2) anatomically map cells expressing candidate receptors using genetic labeling and immunohistochemistry; and (3) functionally investigate candidate receptors and associated circuits using pharmacological and genetic perturbations, coupled with behavioral and paternity assays. Together, these studies will identify the sensory and circuit-level basis of post-mating behavioral plasticity in female mosquitoes. 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 2026 · 2026-05

Modern engineering and economic challenges often involve complex systems that push the limits of traditional approaches, driven by uncertainty and the high dimensionality of real-world data. Scientists in a variety of fields have unprecedented access to massive amounts of data which hinge on complex structures. Decision makers often need to optimize strategies involving large populations full of randomness. This project will address these challenges by developing principled approaches for analyzing and solving complex systems. The project will focus on developing innovations in stochastic control theory and artificial intelligence (AI) algorithms which will inform basic science, technology and economic questions arising in a broad range of disciplines. The results will be disseminated broadly across scientific communities and the general public. The project will also provide training opportunities for graduate students and AI literacy programming at the K-12 level. The project will focus on two interconnected research thrusts to address the aforementioned challenges, centered around the theoretical foundations of stochastic control and games. The first thrust involves studying mean field games (MFGs), which are used to model the macroscopic profile of large interacting systems, unraveling high degrees of freedom of these systems and enhancing tractability of large-scale problems. The goal is to build a quantitative theory for first-order MFGs in response to a surge of interest in modeling large systems with strong signals. Both theoretical and numerical challenges will be addressed, and the project will focus on direct applications in statistical physics and decentralized finance. The second thrust involves advancing generative AI to help mitigate uncertainty in the design of large-scale systems, and provide data-driven insights to identify unforeseen challenges in these systems. The goal of this second part of the project is to build a rigorous mathematical paradigm for diffusion generative models in the context of diffusion model alignment from the perspective of control theory, leading to principled algorithms for generating large, goal-specific data. 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 2026 · 2026-05

Modern technologies depend on understanding how atoms are arranged inside materials. This atomic arrangement determines how materials conduct electricity, withstand extreme environments, store energy, and perform in applications ranging from microelectronics to national defense systems. X-ray diffraction (XRD) and related scattering techniques are powerful tools for revealing atomic structure, yet analyzing XRD data is a complex task, especially since it is usually large in volume. Although thousands of diffraction patterns are generated every year in laboratories and national facilities, the results remain scattered across publications or stored locally without standardized formats, limiting its reuse and slowing scientific progress. There is no public database of experimental powder diffraction data. This project addresses this need. This project will develop DiffAI, an open, community driven platform that will host public experimental powder diffraction data, associated metadata, and provide artificial intelligence (AI) tools for automated analysis. These will enable more accurate structure determination of materials with complicated atomic arrangements, such as quantum materials that may underlie future quantum information technology. By making high-quality diffraction data findable, accessible, and reusable, DiffAI will accelerate and lower barriers for materials discovery. By democratizing access to experimental data and machine learning models, DiffAI will enable efficient analysis of diffraction data and foster collaboration within the global research community. Through open-access tools, student training, and community workshops, DiffAI aims to establish a global standard for sharing and analyzing diffraction data, ultimately driving progress in materials characterization and discovery. This project advances the foundations of scientific cyberinfrastructure in three key ways: 1) A novel, extensible data architecture for experimental diffraction that will combine metadata schemas, JavaScript Object Notation (JSON) based data records, and a public data repository for experimental powder diffraction patterns that supports scalable, persistent storage of heterogeneous diffraction datasets. Persistent Digital Object Identifiers (DOIs), curated releases, and open APIs will facilitate reproducible workflows and long-term sustainability. 2) Automated agentic large language model (LLM) workflows for large-scale data extraction and digitization that identify relevant literature, detect and classify XRD figures, extract labels, and digitize plots into machine-readable formats. The team also plan to develop software tools for more automated data and metadata capture from laboratory instruments and synchrotron X-ray and neutron diffractometers at national laboratories, thereby creating a generalizable blueprint for automated experimental data recovery, an emerging need across multiple scientific domains. 3) Building on prior NSF work, DiffAI will implement domain-adapted AI models integrated into cyberinfrastructure that bridge synthetic training sets with real experimental data for automated XRD data and metadata validation tasks. These will enable more accurate structure determination for complex martials, such as quantum materials. Collectively, these advances will provide a scalable, community-driven cyberinfrastructure element that enables modern, AI-ready diffraction workflows. This award by the Office of Advanced Cyberinfrastructure is jointly supported by the Materials Research Section within the Directorate for Mathematical and Physical Sciences. 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 2026 · 2026-04

The flow of electromagnetic energy through the atmosphere and its interaction with Earth's surface underly an enormous range of weather and climate phenomena. Fluxes of radiation almost entirely determine Earth's temperature, act as a strong constraint on global precipitation and the height of the tropopause, and shape atmospheric motions from the global scale to the cloud scale. The laws of physics for electromagnetic radiation are very well known, particularly for the visible and infrared radiation relevant to Earth, but the application of these laws to the Earth's atmosphere and surface requires complex and intensive calculations, thus the field has relied heavily on computer simulations which are not easy to understand. The last eight years, however, have seen a resurgence of efforts to use the laws of physics in simplified form to build understanding of the influence of radiation on atmospheric phenomena. These advances in understanding have been achieved by embracing the spectral variation, or wavelength dependence, of the flow of radiation through the atmosphere: the world is not black-and-white, nor even grey, but shaped by radiation in all the colors of the rainbow and beyond. In addition to its contributions to the basic science of radiative transfer the project will enrich the workforce by providing support and training to a postdoc or PhD student. This project seeks to build on these recent insights to develop theories and conceptual models in two distinct areas. One set of questions will focus on problems in atmospheric physics: why does the stratosphere cool when carbon dioxide concentrations are increased but warm slightly when the concentration of water vapor increases? How do the tight links between temperature and water vapor mediate the impact of clouds on the radiation leaving the atmosphere at the upper or lower boundaries? How can insights obtained by dramatically simplifying the spectral opacity of greenhouse gases be applied to the absorption of solar radiation? The second set of questions will concern how radiation sculpts atmospheric circulations at all scales by developing a richer hierarchy of simplifications for radiation for dynamical models. The spectroscopy of gases will be replaced by few-parameter idealizations; clouds will be spectrally-uniform and scatter solar radiation in simplified ways; radiative transfer itself may be approximated by focusing on the dominant terms in the relevant equations. These idealized treatments of radiation will be coupled to dynamical models of the atmosphere and used to explore the essential elements of radiation-circulation coupling in a range of increasing complex circumstances. 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 2026 · 2026-03

The award supports U.S.-based graduate students and postdocs to attend and participate in the international conferences and winter school as part of the Thematic Month “Around the Langlands Program” in January-February 2026, at the Centre International de Rencontres Mathématiques (CIRM) Luminy, France. The Langlands Program is a deep and ambitious framework that connects seemingly distant areas of mathematics, primarily number theory, representation theory, algebraic geometry, differential geometry, harmonic analysis etc. The main goal of this thematic month will be to bring together leading experts in the various aspects of the Langlands program, as well as young students and early-career researchers, in order to take a fresh look at recent developments, to introduce students to its main challenges, and to strengthen ties between different mathematical communities that share similar interests. Named after the Canadian mathematician Robert Langlands, who proposed the original ideas in a letter to André Weil in 1967, the Langlands Program aims to establish a far-reaching web of conjectures that relate Galois representations (arising from number theory) with automorphic forms (which come from harmonic analysis). The Program has grown rapidly over the past few decades, and nowadays, there are various manifestations of what is called the Langlands correspondence, ranging from global versions relating global Galois groups to automorphic forms, to local (p-adic and ell-adic) versions relating local Galois groups to representation theory, as well as geometric versions of the Langlands correspondence. A large part of this picture remains conjectural, and it is one of the greatest challenges of the mathematical community to understand what lies behind this phenomenon. The Thematic Month will have weekly themes, as follows: (1) Winter School on Galois representations and automorphic forms; (2) Geometric representation theory in the Langlands program; (3) p-adic aspects of the Langlands program; (4) Geometrization of the local Langlands correspondence; and (5) Relative Langlands and Arithmetic. More information can be found at the program website https://conferences.cirm-math.fr/3498.html. 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 2026 · 2026-03

Axial Volcano is a large submarine volcano located offshore Cascadia. The volcano is currently inflating, and scientists predict that it will erupt in the next few years. This project will deploy seafloor instruments for two years on Axial Volcano. Data from these instruments will be used to monitor the magma chamber before, during, and after the eruption. The results will help in understanding melt migration and eruptions at volcanos. The project will test novel instrumentation and provide research opportunities for a postdoctoral scientist. This project will deploy three new, high resolution seafloor compliance instruments at Axial Volcano for two years. The sites lie over the central magma chamber region that has sourced previous eruptions and span the imaged underlying deep magma conduit. Compliance observations can track shear velocity changes within the magma chamber and constrain changes in melt. The experiment will also facilitate observations of uplift and seismicity by expanding the spatial coverage of the Ocean Observatories Initiative array. The instruments will test large current shields in deep water for reducing tilt induced noise and will include both a differential pressure gauge and an absolute pressure gauge. 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 2026 · 2026-02

The distribution of naturally occurring radioactive isotopes in the ocean can be used to understand the movement of water and particles through the ocean. Measurements of these same isotopes in ocean sediments can, in turn, be used to reconstruct ocean processes back through time. The objective of this project is to measure the isotope beryllium-10 in samples of seawater, ocean particles, and sediment samples from the South Pacific Ocean. The samples were collected on previous oceanographic expeditions, which provides important context for interpreting the new data. The team will use the new data set to test the hypothesis that the distribution of beryllium-10 in the ocean is primarily controlled by particles settling through the ocean water column rather than by ocean circulation. Increased understanding of what controls the distribution of beryllium-10 in the ocean today will improve scientists’ ability to interpret past changes in sediment beryllium-10. The project will support two early career investigators and support development of broadly available earth science education programs. Co-located seawater, particle, and sediment core top samples will be selected from the South Pacific GEOTRACES expeditions GP16 and GP17-OCE. This beryllium isotope data set will provide new constraints on beryllium-10 particle scavenging and oceanic transport dynamics across a wide range of water masses and depositional environments, as well as the impact of these dynamics on the sedimentary beryllium-10 record. The data will be used to examine a variety of first-order questions in marine Be distributions, such as the magnitude and length-scale of boundary scavenging on basin-wide beryllium-10 distributions, potential sensitivity of Pacific beryllium-10 inventories to changes in Atlantic scavenging processes, and the relative impact of local scavenging versus boundary scavenging on sedimentary beryllium-10 deposition rates. In addition, the data will provide points of comparison for dynamic marine Be isotope models and refine scientists’ ability to interpret changes in beryllium-10 fluxes and 10Be/9Be ratios within the sedimentary record. 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 2026 · 2026-02

On 23 November 2025, Hayli Gubbi volcano in northern Ethiopia explosively erupted. The eruption sent ash and volcanic gases nearly 15 kilometers into the atmosphere and disrupting air traffic across large parts of the globe. Eruptions in this region are usually gentle and lava-producing. The explosive nature of this event was very unusual and poorly understood. Volcanic ash fell on nearby communities and contaminated water and grazing lands. Because ash, gases, and surface features degrade rapidly due to erosion and continued gas release, there is an urgent need for rapid scientific study. This team wants to understand how and why explosive eruptions occur in basaltic rift settings. By improving our knowledge of these rare but high-impact events, the project supports NSF’s mission to advance the progress of science and promote public safety. The results will contribute to improved volcanic risk assessment for aviation, infrastructure, and communities in Ethiopia and globally, including in the United States where similar volcanoes exist. This RAPID project will conduct time-sensitive field and laboratory investigations of perishable volcanic products from the 2025 Hayli Gubbi eruption. The team aims to determine eruption magnitude, eruptive style, degassing behavior, and links to regional magmatic processes. Fieldwork will include stratigraphic logging, ash thickness measurements, and systematic sampling of ash and lava deposits, complemented by near-vent volcanic gas measurements. Petrological and geochemical analyses of matrix glass, melt inclusions, and mineral phases using EMPA, LA-ICP-MS, and FTIR will constrain magma storage conditions, volatile contents, and ascent histories. These datasets will be integrated to estimate erupted volume, assess degassing efficiency, and test hypotheses regarding shallow magmatic connectivity between Hayli Gubbi and the Erta Ale volcanic system. The project will provide new constraints on the conditions under which basaltic magmas transition from effusive to explosive behavior, advancing fundamental understanding of volcanism in continental rifts and informing hazard models for similar volcanic systems worldwide.  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 2026 · 2026-01

This project investigates the economic outcomes and trajectories of the fastest growing demographic group in the United States and, in particular, its workforce in science and engineering. To date, there is no consensus on fundamental questions about the labor market experiences and trajectories for this group. Empirical evidence of their labor market outcomes is scattered and mixed despite their high educational attainment. Moreover, this study investigates their labor market experiences disaggregating across gender, nativity, and national origin. The project also yields implications for reductions to barriers to incorporation and intergenerational mobility. This study uses the restricted-access linked longitudinal survey data in NSF's Scientists and Engineers Statistical Data System (SESTAT) and Census Bureau’s Longitudinal Employer-Household Dynamics (LEHD) data and Decennial Census/American Community Survey (ACS) data. This enables it to overcome previous data limitations, which have made it difficult to accurately study the outcomes and trajectories of workers in this demographic group. Because diversity is a hallmark of this population, failing to disaggregate the population results in biased or incomplete narratives. The project adapts two widely used statistical methods—multi-level growth-curve models and group-based trajectory models—to represent individual heterogeneity in earnings trajectories in a population. Findings from these life-course trajectory models reveal economic inequality, such as across workers’ national origin and generation groups, as well as between this and other demographic groups. 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 2026 · 2026-01

All chemical elements are present dissolved in the ocean. Differences in their concentrations and in the relative amounts of their isotopes at various locations and depths are used to understand the workings of the ocean and the geochemical cycles of the planet as a whole. In this project, investigators from three institutions will conduct the first study of the marine chemistry of zirconium isotopes. Pilot data from the team have indicated differences in isotopic composition between seawater and sediments. The proposed work will build on these pilot data by characterizing the zirconium isotopic composition of a comprehensive array of seawater, particle, and sediment samples already collected from the Pacific Ocean. One aim of the work will be to test the idea of whether zirconium isotopes in sediments provide a record of past ocean conditions. The project will support one PhD student and provide research experience for one undergraduate student. The results of this research will be incorporated into class material taught by the PIs and reported in plain language summaries on their research websites and through university press releases. This work will support a diverse, interdisciplinary team of three early career faculty. The team will promote broadening participation in science through targeted educational outreach including the development of educational modules for high school students to spur interest in chemical oceanography and isotope geochemistry among students at an early career stage. Previous work on high field strength elements including zirconium (Zr) and hafnium (Hf) indicated that the ratio of dissolved Zr/Hf varies extensively and systematically with latitude and depth in the ocean. Pilot data obtained by this team on the stable Zr isotopic composition of marine authigenic sediments and seawater has revealed systematic fractionations, but further investigation of this proxy in seawater profiles and marine particulates that intersect major currents is crucial for: i) providing insight into the mechanisms responsible for the isotopic fractionations observed in the ocean; and ii) further developing their application as a potential tracer of paleo-oceanographic processes. The proposed work aims to build upon the pilot observations by characterizing the Zr isotopic composition of seawater, marine particulates, sediments, and leaches in an array of well-characterized samples collected during a previous cruise. This project will apply high-precision, novel non-traditional stable isotope techniques to marine samples to: (1) determine the stable Zr isotopic composition of various endmembers in the ocean (dissolved, particulate, authigenic sediments), (2) probe whether these isotopic compositions correlate with water mass age, (3) test whether the Zr isotopic composition of authigenic sediments faithfully record the water composition they originated from, and (4) test whether adsorption of dissolved Zr onto sinking particulates is the driver of the observed Zr isotopic fractionation in the ocean. 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 2026 · 2026-01

Quantum computing offers the ability to unlock cryptographic advancements that are beyond the reach of existing security technologies. This includes not only strengthening the mathematical foundations of existing cryptographic schemes but also enabling entirely new cryptographic concepts. This project seeks to chart a clearer understanding of the complexity-theoretic foundations that underlie quantum cryptographic systems. The resulting insights will help in precisely characterizing the mathematical assumptions needed for quantum cryptographic primitives. The project also aims to develop new protocols and techniques for secure interaction with quantum information, including ways to verify and protect quantum information. Such protocols could potentially lead to applications of quantum computers and quantum networks in the future. In addition to contributing to foundational science, the project will support educational activities, outreach, and international collaboration. The project undertakes a broad theoretical study of the complexity assumptions and structural underpinnings of quantum cryptography. It aims to clarify the hierarchy of quantum cryptographic primitives and determine which assumptions are necessary or sufficient for constructing various protocols. The research will explore idealized models such as the quantum random oracle model, the common Haar state model, and the Haar random oracle model, in order to probe feasibility and impossibility results. The investigators will also analyze complexity-theoretic reductions and separations between different cryptographic primitives, and employ recently developed unitary complexity theory to better understand their security. A further goal is to design new quantum cryptographic protocols, including interactive and zero-knowledge protocols for preparing quantum states, and to explore notions such as proofs of quantum knowledge. By pursuing these directions, the project seeks to build a coherent, complexity-theoretic framework for quantum cryptography. 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 · 2025-10

The transport of magma is a fundamental process in volcanic eruptions, crustal extension and the creation of new crust on earth. Dikes and sills are magma-driven fractures that often record past evidence of this process. Over the past two decades, observations of magma transport in the crust have highlighted the need to better understand how magmatic and tectonic processes interact to emplace dikes and sills on earth. In this study, PI Scholz will conduct a combination of numerical and analog modeling to address the fundamental dynamics of the transition from dike to sill and from sill to dike in extensional tectonic settings, like the Reykjanes Peninsula in Iceland. These models will lend insight into how and where magma is stored before volcanic eruptions and into the triggering mechanisms of dikes. A better understanding of these factors is at the core of forecasting the timing of volcanic eruptions and, consequently, volcanic risk mitigation efforts. Furthermore, the project will incorporate hands-on opportunities to engage local K-12 audiences through existing outreach avenues at LDEO and mentor an undergraduate student researcher in the lab. This project will address two key questions. Does the change in stress associated with dike opening influence magma flow into sills and dikes during plate spreading and extension? Are dikes at spreading centers and continental rifts more likely to be triggered by earthquakes driven by lithospheric extension or by magma overpressure caused by pulses of magma rising from below? PI Scholz aims to demonstrate that the stress field associated with dike opening could induce sill formation using analog gelatin tank experiments that reflect the structure of spreading centers. The results from these analog experiments will be combined with numerical models for repeated dike and sill injections to understand how intrusions impact the local stress field and, consequently, the depth of sill formation during extension. To address the second question, PI Scholz will extend an existing geodynamic model for extensional margins to explore the relative importance of earthquakes and the buildup of magmatic overpressure on the initiation of dike events. By integrating these model results with existing geophysical datasets, this work will produce a novel model for the frequency of dike events in extensional settings. 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 · 2025-10

This award supports two years of funding for the U.S. GEOTRACES Project Office to coordinate and support the activities of investigators at U.S. institutions who are fulfilling the GEOTRACES mission, specifically: “To identify processes and quantify fluxes that control the distributions of key trace elements and isotopes (TEIs) in the ocean, and to establish the sensitivity of these distributions to changing environmental conditions.” To fulfill this mission, and achieve benefits derived from doing so, GEOTRACES is making observations spanning a global array of ocean sections in close collaboration with modeling and synthesis, and supported by a solid foundation of ongoing intercalibration of analytical methods. A program-wide data management system ensures that the results of the program will live long after the program is completed. The US GEOTRACES program is at a time of transition, having recently completed its final section cruise and looking to embark on targeted “process” studies while emphasizing synthesis of GEOTRACES findings. This award provides support for Project Office personnel, for workshops and activities contributing to broader GEOTRACES synthesis, and for activities that facilitate the use of GEOTRACES products by the broader scientific community as well as learners, educators, policy makers, and other stakeholders. U.S. GEOTRACES has completed its prioritized sequence of ocean sections, resulting in over 500 papers, with many more to come. Data are archived at the Biological and Chemical Oceanography Data Management Office and incorporated into an international database. Data are freely available for anyone to use, as are software tools to facilitate its use. This award will enable the U.S. GEOTRACES Project Office to continue coordinating the implementation of future process studies and to provide a research framework from planning through sampling to data synthesis. At this phase of the Program, a major priority is synthesis, that is, to interpret measured distributions of TEIs with the goal of identifying the principal sources of TEIs in the ocean, quantifying the rates of processes that regulate TEI supply and removal, and assessing the role of TEIs in biogeochemical cycles of carbon and major nutrients. This often requires going beyond disciplinary and national boundaries. The Project Office will continue to promote and facilitate these activities through a combination of large synthesis meetings and smaller synthesis working groups. The Project Office will engage in a multipronged effort to broaden the use of GEOTRACES findings, including: 1) distributing newsletters to the US oceanographic community and identifying opportunities within GEOTRACES to encourage the use of its data; 2) training early career scientists and giving them the tools to lead similarly large programs in the future; and 3) leveraging the selection of GEOTRACES within the UN Decade of Ocean Science to assist emerging oceanographic programs establish policies and procedures aimed at achieving the best possible data management 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 · 2025-10

Non-technical description: The goal of this project is to build and steer swarms of micron-scale magnetic colloidal particles that come together and move cooperatively through complex environments, much like schools of fish, flocks of birds, or swarms of insects. These swarms are activated by a time-varying magnetic source (for example, an electromagnet or a moving permanent magnet) which functions as an external remote controller. The magnetic controller can direct swarms to propel through fluids, maneuver over surfaces and around obstacles, detect and respond to changes in their surroundings, and carry passive cargo. This project aims to advance the field of magnetic swarms by integrating large computer simulations, theoretical modeling, and experimental approaches within a cohesive framework. Mastering life-like swarm behavior could enable miniature ARMS robots that deliver medicine inside the body, inspect subsurface pipelines, or remove contaminants from water supplies. By opening new frontiers in materials science and programmable matter, this project advances the nation’s health, prosperity, and security while strengthening technological leadership. This project will also provide K-12, undergraduate, and graduate students with interdisciplinary training in computational and experimental techniques for materials science, physics, and engineering to develop our domestic workforce, improve public scientific literacy, and stimulate engagement with science and technology. Technical description: While magnetic swarms capable of dynamic reorganization have been demonstrated, a systematic approach to designing swarms with increasingly sophisticated functions in porous environments and unbounded 3D fluids remains a challenge. Large-scale simulations will capture the coupled magnetic, hydrodynamic, and contact interactions that drive collective motion across multiple length and time scales. Analytical theory will translate these data into design rules, while inverse-design algorithms will search efficiently for particle shapes, magnetic moments, and field protocols that enable adaptive aggregation to move through complex structures. Lithographically fabricated and chemically synthesized particles will test these predictions; high-speed imaging, particle tracking, and force mapping experiments will measure swarm structure, flow fields, and cargo transport efficiency. By combining computational analysis with experimental methods, swarm functionalities for advanced applications, such as adaptive organization, precise navigation, and targeted cargo transport in complex environments will be expanded. These advancements will create a foundation for future applications of colloidal swarms in sensing and delivery, turning theoretical insights into practical outcomes. More broadly, the proposed methods to accelerate swarm design will benefit other active material systems where flows of energy, matter, and information animate material structures to enable life-like capabilities. 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 · 2025-10

The Institute of Electrical and Electronics Engineers (IEEE) Symposium on Foundations of Computer Science (FOCS) is one of the premier annual research conferences that cover the breadth of theoretical computer science. It is a conference of very long standing that has continued to play a formative role in the field; it is also at the leading edge of connections made to other areas. This project aims to increase the impact of this conference on students and postdoctoral researchers by encouraging and enabling their participation, especially in cases where travel expenses would otherwise preclude their attendance. Concretely, this project assists US-based students and postdoctoral fellows in attending the 2025 Annual FOCS conference, sponsored by the IEEE Computer Society Technical Committee on Mathematical Foundations of Computing (TCMF). The coming FOCS will take place in Sydney, Australia from December 14-17, 2025. As computing becomes ubiquitous, it is crucial to expand the participation of young scholars in cutting-edge research. FOCS has served as one of the most important venues for groundbreaking research in theoretical computer science – and, increasingly, as a key ambassador to other areas within and outside computer science. We anticipate the conference presentations and discussions will expose students and postdoctoral researchers to a broad set of fundamental questions and ideas. We expect the community to benefit in turn, as such junior researchers have contributed substantially to the growth of theoretical computer science over the years. With an emphasis on junior scholars in need, supporting the participation of junior researchers in the stimulating exchange of ideas benefits all conference attendees. As the tools and techniques from theoretical computer science (such as novel models, algorithms, impossibility results, and unexpected connections in data science and machine learning) are becoming vital to several domains inside and outside computer science, it is anticipated that such broader participation of junior researchers will benefit society at large. 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 · 2025-10

This Faculty Early Career Development (CAREER) award enables contribution of new knowledge related to the manufacturing process for thermoset composite structures. Thermosets are advanced polymers that can be reinforced with high-strength fibers to create lightweight composite materials with complex architectures. Their manufacturing process involves chemical curing reactions, heat generation, and heat conduction that result in a change of phase from a liquid polymer mixture to a cross-linked solid structure. Stresses built up during the process can cause very small cracks to be formed, which can contribute to subsequent failures in the structure. Extensive curing-induced microcracking is a key technological challenge in the production of critical structural components for space exploration, wind energy production, and current and future transportation. The manufacturability of the next generation of high-performance, lightweight structures depends upon fabricating damage-free complex composites with enhanced mechanical properties. This award supports fundamental computational and experimental research to provide the necessary knowledge for developing crack-free thermosets, optimizing composite performance, and reducing the time and cost to create better composite parts. This research program will be integrated with educational and outreach activities, including developing an e-learning platform with engaging learning activities, K12 summer programs, and internships that aim to broaden the participation of underrepresented groups in research and positively impact engineering education. The research goal of this project is to reveal the fundamental mechanisms of curing-induced microcracking in thermoset composites. This project will establish process-property relationships to predict curing-induced damage mechanisms in thermosets across the micro and macro scales, which will enable new manufacturing capabilities. Knowledge generated from this research will allow manufacturers to tailor their processes to prevent curing-induced damage. This project will test the hypothesis that layer/layer and fiber/matrix property mismatches and resin shrinkage cause microcracking when the material is processed below its glass transition temperature and after gelation, depending on the resin viscosity and toughness. Advanced multiscale process modeling techniques that account for thermal gradients, resin exothermic reactions, mismatch in thermomechanical properties, shrinkage, and residual stresses will be implemented. A new time-independent characterization technique for determining material properties at intermediate degrees of cure based on off-stoichiometry polymer proxies will be tested. This novel approach to learning constitutive relations of thermosets during curing will be used to identify and quantify viscoelastic and viscoplastic resin properties during manufacturing. In-situ testing during curing will aim to validate the approach for three material systems, including 3D woven textiles, bonded adhesives, and thermosets for additive manufacturing. The planned result is an experimentally-validated physics-based multiscale process modeling framework to design and optimize enhanced composites which can help the composite industry by providing a missing link between material, manufacturing, and properties in order to prevent microcracking. 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 · 2025-10

As the Internet of Things and the emergence of smart materials proliferate, the demand for inexpensive, long-lasting, and efficient electroluminescent devices will continue to increase. Perovskites are solution-processible and Earth-abundant crystalline materials exhibiting high efficiency, attracting significant industrial attention. We have recently demonstrated simple light-emitting devices constructed from a single perovskite layer termed light-emitting electrochemical cells, which utilize battery-like charging to achieve high brightness and efficiency. Taking advantage of their simple architecture and facile construction, we will leverage robotic fabrication to rapidly produce arrays of varied perovskite thin film devices for optimized performance. To begin, we will precisely tune the lithium-ion charging of these devices for enhanced efficiency and stability. Next, we will robotically fabricate films and devices with a continuous range of colors, green through red, by tuning the chemical composition of the perovskite crystals. Finally, we will utilize robotic fabrication to tailor the crystal structures of these films, resulting in crystalline sheets that exhibit high-efficiency blue emission. This work will train undergraduate and graduate researchers for competitive careers in semiconductor technologies. The project will develop new technologies suitable for commercialization and provide insight into existing approaches. The program will create a new instrument for high-throughput device fabrication and testing. Finally, we will engage the public at large with an interactive color mixing and perception laboratory. A surge in the efficiency of perovskite light-emitting devices (PeLEDs) and light-emitting electrochemical cells (PeLECs) offers the potential for high-efficiency, low-cost, long-lived, and mechanically flexible light-emitting devices. Recently, we have demonstrated perovskite light-emitting electrochemical cells, devices that leverage differential ion motion to produce high performance in a simple, single-layer device. Through robotic fabrication of thin films and devices, we will achieve the following: -Robotic fabrication will enable fine-tuning of lithium doping concentration for optimized performance. Lithium doping has been shown to significantly improve the lifetime and performance of this class of devices. It is hypothesized that the lithium cation and counterions serve as surrogate mobile ions in these devices, replacing the parent ions of the perovskite. Additionally, these additive ions fill traps and passivate voids, thereby limiting nonradiative energy losses and stabilizing the perovskite crystallites. -Yellow to deep red electroluminescence will be achieved by robotically balancing the I− /Br− concentration ratio. Commercial LEDs have suffered from wide spectra with impure colors, particularly in the green-to-orange portion of the color gamut. Color tuning with a mixed iodine/bromine solution offers a pathway to address this, but the system is highly concentration dependent. Robotic assembly will enable consistently color-tuned and efficient devices for commercially underperforming LED colors. -3D-2D films will be robotically produced that incorporate small stacking number quasi-2D phases. The most promising perovskite devices, those with high efficiencies and world-leading lifetimes, have utilized multi-phase blends of perovskites. Two-dimensional phases blended with standard three-dimensional perovskite (3D-2D devices) are an emerging area for achieving breakthrough performance through quantum confinement in quasi-2D phases, utilizing spacer and coupling additives. This approach creates both an exciting opportunity and a significant challenge: Several 3D-2D fabrication methods are possible due to spacer and coupler molecules and their relative concentrations, but the throughput of traditional fabrication is highly rate-limiting. Robotic assembly will greatly boost the exploration of this exciting experimental landscape. 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 · 2025-10

The goal of this research project is to build mathematical foundations for reasoning about the behavior of modern machine learning systems. Foundations are needed to direct future developments in Artificial Intelligence (AI) research, and also to diagnose and remedy problems that arise in existing AI systems. This project specifically focuses on how AI models represent data. For example, many language models are "trained" using vast collections of text from books, websites, and other sources, but these texts are not "stored" in the model in the same way that documents are stored on a computer. Rather, the texts are transformed in a way that seems to facilitate their use for a variety of tasks, ranging from generating code for a website to solving mathematical word problems.These representations, however, are not perfect, and they also seem to lead to embarrassing errors committed by AI models, such as miscounting the number of Rs in the word STRAWBERRY. Developing a mathematical theory of the representations used by AI models will help demystify how the models perform these tasks and reveal fundamental limitations that result in errors. The theory will also guide the development of next-generation models that go beyond the limitations of current models. Feature learning is a key ingredient in the success of modern machine learning systems, and thus its understanding is a essential in any theory of deep learning. The aims of this project are as follows. The first is to develop general principles of how features emerge by building on well-established mathematical structures, such as low rank structure and circulant structure. The second is to study these principles in analytically tractable and practical architectures--such as two-layer multilayer Perceptrons, kernel methods, and transformers--in the context of specific data frameworks/inference problems such as multi-index models and modular arithmetic. The third is to develop new architectures that are potentially viable alternatives to neural networks that are currently in widespread deployment, thereby mitigating risks of over-reliance on specific technologies while suggesting directions for improving or controlling current architectures. 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 · 2025-10

Extended reality (XR) technologies have shown significant promise in increasing user engagement and skill acquisition in a variety of domains. The goal of this project is to accelerate adoption of innovative XR applications for rehabilitation, for example, to enhance user experiences in treatments to improve motor function for diseases with motor disabilities. This project develops immersive XR exercise environments that enable users to move and interact in 3D space with each other and with virtual elements. A key goal is to make the experience of rehabilitation more enjoyable and effective by incorporating social interactions between remote users that can boost engagement and skill acquisition. The technology also allows clinicians to guide and interact with their patients remotely. To create a virtual environment that users experience as fast and seamless, this project develops novel approaches to the underlying networking infrastructure needed to run the application. Collaboration with industry partners will support technology adoption for XR-enabled rehabilitation technology and for the XR industry. Realizing multi-user, geo-distributed XR technology is challenging due to stringent motion-to-photon latency requirements for good user experience. Current wide-area Internet routing and cellular wireless management are one-size-fits-all across applications, hurting latency. A key insight of this project is that not all types of XR traffic require uniformly low latency; instead, the project enables prioritization of delivery of important XR traffic while intelligently managing lower-priority XR traffic. The project’s core technical contributions are an adaptive XR application with new delivery mechanisms, leveraging Internet path selection on the testbed, and programmable wireless resources using Open Radio Access Networks (Open RAN) and xApps. Performance of the XR environment will be assessed via user experience and key network performance indicators. The expected outcome is achieving key performance targets currently deemed impossible on today’s Internet by demonstrations of network-supported, multi-user XR technology for rehabilitation at different sites. 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.