University of Maryland, College Park

NSF Awards · FY 2026 · 2026-01

Tidal marshes are one of the most biologically productive ecosystems on earth, supporting critical ecosystem services including coastal storm defense, fisheries support, and unique biodiversity. Yet, tidal marshes are severely threatened by coastal flooding. Trapped tidal floodwaters break-up marsh meadows (“ponding”), and cause remnant marsh grass to become sparse and stunted. Healthy meadows are needed by marsh-breeding birds to prevent nest flooding, an increasing cause of nestling mortality leading to species declines as fast as 9% per year. To save marsh birds, practitioners have proposed a novel technique to drain trapped floodwaters by creating shallow channels (“runnels”). However, effective methods are still being developed, and whether this approach to restoration will work for tidal marsh birds is unknown. This project will examine marsh bird responses to ponding, measure ecosystem responses to runnels, and provide design metrics and restoration targets for runnels. In 2025, a National Audubon Society-led partnership (“Marshes for Tomorrow”) set an ambitious goal to restore thousands of acres of tidal marsh in Maryland by 2035. Using results from this project, Towson University, University of Maryland, and Audubon will take action toward this goal through evidence-based restoration. The project will produce technical protocols for practitioners, make public outreach presentations, and train undergraduate and graduate students in ecological science and conservation practice. The project will also engage with local communities to choose restoration sites, and share science-based management options with landowners to reduce flooding of their marshlands. This project investigates relationships between ecosystems, landscape-scale patterns of fragmentation (ponding), and consequences for tidal marsh bird persistence — integrating across spatial and temporal scales to link ecosystem function to bird population demographics. The study will focus on two tidal marsh endemics, the rapidly declining Saltmarsh Sparrow, and the Seaside Sparrow. Surveys and experiments will be conducted in twelve tidal marshes in the Maryland-portion of the Chesapeake Bay. This project includes three technical approaches: 1) large-scale habitat and breeding bird surveys using field studies and remote-sensing, 2) an ecosystem-scale restoration experiment monitored using field surveys and biogeochemical analyses, and 3) microhabitat studies using field-based nest observations and mesocosm experiments on plant and soil recovery. The project will test the hypothesis that tidal marsh bird densities and nesting success decline non-linearly as ponding increases (exploring the presence of tipping-points) and will derive habitat indices from remote-sensing data to direct site-selection and set target metrics for habitat recovery after restoration. Audubon will conduct hydrologic restoration at an expansive, moderately fragmented marsh that researchers will use as a whole-ecosystem experiment with a before-after-control-impact design. By assessing variation in water, soil, and plant responses to runnels the project will identify specific characteristics favorable for restoration. Spatial and temporal patterns in marsh bird responses to pond-fragmentation, ecosystem responses to restoration, and subsequent implications for habitat quality will inform basic science on resilience and recovery. 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

With the support of the Macromolecular, Supramolecular and Nanochemistry program in the Division of Chemistry, Professor YuHuang Wang of the University of Maryland, College Park, is developing innovative chemical strategies of using programmable DNA templates and laser-based refinement to guide the precise placement and structural control of organic color-centers. Organic color centers are atomic-scale defects that can be engineered onto the surfaces of single-walled carbon nanotubes, which are tiny cylinders made entirely of carbon atoms. Each color center can emit photons—the elementary particles of light—one at a time, even at room temperature. These sources of “quantum light” are essential building blocks for future technologies in secure communication, bioimaging, and molecular sensing. The project will provide students with hands-on experience in quantum nanochemistry and incorporate nanoscience principles into teaching and educational initiatives with broader public. The interdisciplinary nature of the project, which combines chemistry, photophysics, and nanoscience, will provide unique training opportunities to cultivate the next generation of scientists and engineers. In this project, a DNA-programmed photochemical platform for the deterministic synthesis of organic color centers (OCCs) on single-walled carbon nanotubes will be developed. Spatially encoded DNA scaffolds will define reactive sites along the nanotube surface with nanometer precision, enabling the formation of isolated OCCs and OCC pairs with controlled spacing and orientation. This strategy could provide a level of control that is unattainable with currently existing methods. The prepared OCC pairs will then be refined through localized photothermal annealing to tune their atomic structure and coupling strength, guided by in situ hyperspectral imaging capable of resolving OCCs at the single defect limit. This approach will allow for fundamental studies of OCCs with controlled defect-defect interactions, offering a scalable synthetic pathway toward quantum materials with programmable optical properties. 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 · 2025-12

The planet Earth consists of three main layers: the metallic core, the silicate mantle, and the crust. Earth’s outermost layer is a thin shell of solid rock where life flourishes. The crust under oceans, called oceanic crust, is thinner and denser, and continental crust that forms continents is thicker and lighter. The mantle lies beneath the crust and is approximately 2,900 km thick comprising 67% of the total mass of the planet and 84% of its volume. The mantle is mostly solid, but it can flow slowly over long periods of time due to high temperatures and pressures, driving plate tectonics. The Earth's innermost layer, the core, is extremely hot and dense. While the core has been a near-closed system since the time it formed, the crust and the mantle represent a very dynamic and ever-changing system constantly interacting through the continuous transfer of mass and heat. One of the most fundamental challenges in the Earth sciences is reconstruction of the chemical evolution of the crust-mantle system. Samples of some of the remaining well-preserved vestiges of the early rock record, the invaluable time capsules that may shed light on the Earth's distant geological past, have been assembled by the PI of this project over his entire scientific career; these samples will be interrogated in this project using state-of-the-art isotope and geochemical tools to help decipher the evolution of the early Earth's crust-mantle system. The work for this project will include creating a comprehensive model for the timing of formation of the continental crust and complementary chemical evolution of the mantle, significantly advancing our understanding of early Earth’s history. Therefore, it will have relevance to the long-debated question of how terrestrial planets formed and evolved and will ultimately improve our understanding of the modern Earth. This project will involve training students for their future scientific careers. This research project seeks to constrain the history of continental crustal growth and mantle depletion in the early Earth by obtaining estimates on the timing of extraction, absolute volumes, and relative proportions of continental and oceanic crust in the Archean. A novel approach of combining lithophile element isotope (Pb-Pb, 146,147Sm-142,143Nd, and 176Lu-176Hf) and trace element (e.g., rare earth elements, Nb, U, Th, Hf) abundance data obtained using the state-of-the-art analytical techniques for a global set of thirteen komatiite-basalt systems ranging in age between 3.6 and 2.7 Ga and located on five continents, will be utilized. These komatiite-basalt suites were selected because: (a) their emplacement ages have been robustly determined using several independent geochronometers, (b) a large set of high-precision short- and long-lived radiogenic isotope and lithophile trace element abundance data are available for most of them from the previous and current studies of the PI, and (c) these komatiite-basalt suites have been shown to have escaped contamination with the material of earlier-formed upper continental crust and their compositions were mostly unaffected by secondary alteration. Using the combined isotope and trace element data both available from the previous work of the PI and obtained in this study, the time-integrated Sm/Nd, Th/U, and "true" Nb/U ratios of their mantle sources will be calculated. Using these data, the absolute volumes and relative proportions of continental and oceanic crust extracted for each time frame defined by the ages of the studied komatiite-basalt systems will be estimated, and a well-constrained model of the history of continental crustal growth and complementary mantle depletion in the Archean will be generated. While working on this project, the PI will continue supporting the integration of research and education via training students, postdoctoral researchers, and visiting scientists. The PI and the undergraduates involved in this project will develop a display for the annual Maryland Day event highlighting the results from this project, presenting current ideas about early Earth history to the public and future students. This project will make a significant contribution towards building the still extremely limited high-precision radiogenic isotope and trace element database for the Earth’s earliest mafic-ultramafic rock record, and this database will be made available to the broader scientific community for the future research and development efforts. 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 · 2025-12

The natural world is governed by wave equations: the electricity on a circuit board, the light in fiber-optic cables, the elementary particles inside atoms, and even the black hole in the center of the galaxy all propagate by wave dynamics. Though ubiquitous, wave-type equations are far from well-understood. The goal of this project is to understand how waves are affected by interference with themselves or with their environment. The research seeks to learn when and why some waves disperse, other waves persist, and still others collapse. Knowing how waves behave drives technological progress - smaller microchips, faster data transmission, and deeper insights into the fundamental physics of the universe. The project provides research training opportunities for undergraduate students, graduate students, and postdoctoral researchers. The investigator studies the long-time dynamics of solutions to nonlinear wave and dispersive equations, focusing on equations that admit topological solitons, which are used to model the physical phenomena described above. Solitons are localized solitary waves with a nontrivial topological invariant. They were introduced by Skyrme in the 1960s as candidates for particles in classical field theories. They have properties required from a particle in classical mechanics - one can define their position, momentum, and energy - and viewed from a distance, configurations of multiple solitons resemble systems of interacting particles. The investigator's work on multi-soliton dynamics makes this connection with classical mechanics explicit, reducing the dynamics of strongly interacting solitons to underlying n-body problems for their positions, momenta, scales, etc. A guiding principle in the analysis of soliton dynamics is the Soliton Resolution Conjecture, which predicts that generic solutions decompose near the final time of existence into a superposition of finitely many solitons and a term capturing the radiation, often a solution to the underlying linear equation. The investigator will work towards proving the conjecture in certain settings and going beyond it in others by considering three categories of problems: (1) the soliton resolution conjecture for evolution equations without symmetry assumptions, starting with the harmonic map heat flow in two dimensions, which is a long-standing open problem; (2) the unique continuation problem for singular nonlinear waves past the blow-up time; and (3) the question of giving asymptotic descriptions of multi-soliton solutions and their collisions. 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 · 2025-11

Deep inside every drop of water lie tiny signatures from rare forms of hydrogen and oxygen, known as isotopes. These isotopes contain different numbers of neutrons and act like nature’s recorder, revealing where water vapor came from and how it traveled across the globe and formed precipitation. Sensitive to both temperature and humidity along their pathways, these isotopes are stored in natural archives such as ice cores, lake beds, and cave formations. Scientists can read them like time capsules, unlocking stories of Earth’s past climate: how hot it was, how much it rained, and how climate patterns shifted over thousands of years. These records also help scientists evaluate and improve water cycles in climate models, offering insights into how our planet may change in the future. However, adding isotopes into today’s complex climate models requires cutting-edge scientific and engineering expertise and vast amounts of computing power. The goal of this project is to build a smart shortcut based on machine learning: an “emulator” that can predict water isotope patterns from climate variables in existing climate simulations quickly and efficiently. This is a powerful step forward for climate science, hydrology, and understanding our planet’s past and future. The project fosters interdisciplinary collaboration between climate scientists and AI experts. Its success could lead to new ways of modeling other passive tracers in the Earth system. The project products such as code and data will be open to the community, and results will be shared through university courses, training programs, and K-12 outreach at NSF NCAR, Pitt, and UMD. Using simulations from fully-coupled global climate models (GCMs) with isotope capabilities, such as the isotope-enabled Community Earth System Model (iCESM1), this project aims to build a machine-learning-based emulator that learns a mapping from climate fields to water isotope fields. This mapping can then be applied to other GCMs that lack built-in isotope capabilities, enabling cost-efficient generation of isotopic outputs. Scientifically, the project will improve understanding of the leading drivers of isotopic variability, enhance model–data comparison using both modern and paleoclimate observations, and support isotope-enabled climate model development. It also contributes to ongoing (paleo)climate data assimilation efforts in the broader community, where the lack of isotopic prior simulations has been a limiting factor. 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 will support research that looks to derive a fundamental understanding of molecular interactions and toughening mechanisms in inorganic minerals for tailoring tough and durable structural materials. A key innovation is anticipate to lie in turning ubiquitous calcium carbonate into a monolithic binder as a potential alternative to cement and concrete, rather than a raw material thermally decomposed in traditional cement manufacturing. Towards this end, novel synthesis and strengthening pathways will be explored to address the fundamental challenges in constructing continuously structured inorganic monoliths and tackling the poor fracture toughness and low tensile strength of crystalline minerals. By redefining the synthesis process and improving the properties of materials and structures in a truly sustainable and cost-effective way, the anticipated project outcomes could ultimately shed light on multiple research areas and industrial sectors, including civil engineering, materials science, mechanical engineering, advanced manufacturing, and the utilization of abundant and renewable resources with long-term economic and environmental benefits. The research efforts will be integrated with educational activities, including the BRIGHT (Build Resilient, Innovative, and Green Homes on Terra), Science Playground, and SEED (Sustainability Exploration, Engagement, and Discovery) programs, to offer interactive, hands-on learning experiences for future engineers and scientists. The principal hypothesis of this research is that inorganic carbonate minerals can be turned into polymerizable phases capable of forming monolithic binders if stable clusters with controlled sizes can be tailored and regulated to trigger non-classical nucleation and crystallization. This hypothesis will be tested by computationally and experimentally investigating polymer-like precursors, refining reaction pathways, and regulating molecular-scale interactions. Inspired by the mineralization pathways in living organisms, the unique non-classical strategy looks to provide a thermodynamically favored pathway for tailoring mineral-based binders under ambient conditions by bypassing the critical free energy that must be overcome in conventional approaches. To enhance toughness and tensile strength, multiple bio-inspired toughening mechanisms, including copolymerization with organic monomers to tailor hybrid molecules, incorporation of coherent nanodomains to trigger pre-strained crystal lattices, and construction of packed and aligned nano-reinforcement, will be tailored and integrated across multiple length and time scales. The fundamental knowledge and insights gained from this project look to advance the reimagining of structural materials design by unlocking a pathway that is prevalent in nature yet rarely replicated through artificial synthesis, offering transformative material solutions for future civil infrastructure. 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 CAREER project focuses on building text generation systems that interact with people to improve their writing and also help them learn to write. Such “machine-in-the-loop” writing assistants are potentially transformative technologies for improving the writing quality and productivity of human authors, as well as providing new tools for writing pedagogy through cyberlearning applications. However, they have been relatively underexplored by the natural language processing community due to major difficulties in modeling, evaluation, and data collection. The technologies developed in this project address these challenges by (1) developing platforms that leverage existing online author communities to enable the design and evaluation of machine-in-the-loop writing assistants; (2) advancing text generation modeling to improve the quality of generated text; and (3) enabling assistants to rewrite and reorganize human-authored text through developments in automatic paraphrasing. In addition to aiding authors in online communities, the writing assistants developed through the project are deployed in K-12 classrooms to advance writing pedagogy. The project incorporates undergraduate students, including those outside of computer science, in natural language processing research, and provides significant outreach to underrepresented minorities. To make meaningful progress on the development of machine-in-the-loop writing assistants, the project includes a collaboration with Protagonist Labs, which runs online platforms for collaborative storytelling in both creative and pedagogical settings and already has incorporated systems built by the investigator’s team into user-facing interfaces. User interaction on such platforms allows fine-grained evaluation of novel text generation methods, which include neural language models that integrate context compression, context retrieval, and discrete latent variables into the generation process to improve overall coherence and relevance. In addition to producing new text, fully-featured writing assistants must also be able to rewrite user text into a specified form, such as a target writing style or suitability for a target audience. To this end, the project introduces new paraphrase generation models at a variety of units of text, including phrases, sentences, and paragraphs. This research effort aims to spur research into interactive text generation systems, and as such its outputs will include publicly-released pretrained models and open-sourced code. 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

AI Coding Assistants have revolutionized software development by significantly boosting developer productivity. By 2028, it is estimated that 75% of software engineers in enterprises will rely on AI Coding Assistants. These assistants are powered by Code Large Language Models (Code LLMs), which are already integrated in software development environments to complete partial programs and generate code from natural language instructions. However, Code LLMs can produce vulnerable code, raising serious security concerns. Thus, it is critical to ensure the security of code generated by Code LLMs. This project seeks to understand the secure coding capabilities of Code LLMs and to develop techniques to ensure both security and correctness of their outputs. The project’s novelties are establishing the foundation for multi-objective security evaluation of Code LLMs and advancing all facets of the Code LLM security ecosystem to support secure and reliable code generation. The project's broader significance and importance are enhancing the security and reliability of AI-driven software development, empowering millions of developers and tens of thousands of organizations to strengthen critical software systems, and promoting societal and educational impact through student mentoring and training. The objective of the project is to design, develop, and implement the next generation of secure Code LLMs that can be used in realistic coding scenarios. The project is organized into three key research thrusts. First, the project develops multi-objective benchmarks and metrics to thoroughly evaluate Code LLMs in a variety of practical coding scenarios at different difficulty levels, including foundational programming, real-world project development, and instruction following. Second, the project develops semantic-aware decoding algorithms for Code LLMs to generate secure and correct code by enforcing security constraints and ensuring code quality simultaneously. Third, the project designs and implements semantic-aware pre-training techniques to learn secure coding practices while maintaining the performance of general AI coding abilities. The project will release the resulting datasets, methods, and scientific findings to the research community. 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

High-resolution geoscience data are essential for understanding and predicting extreme weather events, yet producing such data remains a major challenge due to limitations in observational infrastructure and computational cost. This project introduces a transformative AI-based framework to overcome these barriers by generating high-fidelity, physically consistent, and uncertainty-calibrated geoscience data. These enhanced datasets will empower better decision-making in disaster preparedness, emergency response, and infrastructure planning. The project’s broader societal impacts include advancing tools for tropical cyclone prediction and wildfire detection, training a new generation of interdisciplinary scientists in AI and geosciences, and releasing open-source software for broad accessibility. By integrating explainable AI with physical principles and expert knowledge, the research aims to improve public trust in scientific models and provide actionable insights for meteorologists, policymakers, and emergency managers. Outreach efforts and mentoring initiatives will promote participation in STEM and foster the development of future leaders in climate resilience and AI for natural hazards. This project develops a next-generation generative downscaling framework that combines diffusion-based generative models with physical constraints, domain expertise, and probabilistic uncertainty quantification. Key innovations include physics-guided loss functions to enforce geophysical realism, text-prompted guidance from expert knowledge for tailored downscaling, and conformal prediction techniques to provide rigorous confidence intervals for model outputs. The approach will be validated using diverse, high-impact datasets such as IMERG, CMORPH, and radar observations, with specific applications focused on improving precipitation estimation in tropical cyclones and enhancing wildfire risk detection. By unifying advances in machine learning, statistical modeling, and atmospheric science, the project will establish a new foundation for trustworthy AI-driven downscaling and support critical scientific and societal needs in a changing climate. 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

Immersed granular materials such as river and seabed sediments and industrial slurries consist of small particles dispersed in a liquid. When these particles stick together, the bulk behavior of the slurry becomes highly complex. These cohesive effects determine whether a riverbed erodes, a coastal slope collapses, or an industrial process is successful. Despite its importance, predicting the bulk behavior of cohesive immersed grains is difficult. This project addresses the challenge of predicting how microscopic adhesive forces between individual particles control the large-scale flow of immersed cohesive granular materials. The resulting knowledge will help improve models for underwater sediment transport, water treatment facilities, and industrial slurries. The project will combine a novel laboratory approach, utilizing controllable "sandcastle-like" bonds between particles with advanced particle-resolved numerical simulations that connect the particle and bulk scales. The project also includes a strong educational component that will provide research and training opportunities for high-school, undergraduate, and graduate students to develop the next-generation scientific workforce. The goal of this project is to develop a quantitative framework connecting particle-level cohesion to the macroscopic flow and rheology of immersed granular materials. The project will integrate laboratory experiments with particle-resolved numerical simulations. The experiments will directly measure the adhesive force induced by capillary bridges between immersed particles, characterize the dynamics of cohesive particle clusters using high-speed imaging, and quantify bulk flow properties including yield stress in canonical configurations like granular collapses and rotating drums. These results will be used to develop and validate a cohesive force model for the simulations, which will then probe the microstructural origins of the bulk response of the cohesive granular material. The primary scientific contribution will be the formulation of data-driven, continuum-level constitutive laws for cohesive immersed granular flows. This work will develop a validated computational tool and provide a foundational understanding to improve predictive models for cohesive sediment transport and other applications. 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 traditional approach to new materials discovery and optimization emphasizes perfect crystallinity, minimization of disorder, and the avoidance of complex symmetries, structures, and chemical compositions. High entropy materials, however, embrace a random multi-element solid solution distribution with large configurational entropy to dominate the enthalpy of formation, leading to optimized materials properties. This project aims to develop a new subfield, High Entropy Quantum Materials (HEQMs), to elucidate the potential of configurational entropy for expanding the landscape of quantum phenomena. Utilizing the power of configurational disorder in stabilizing exotic phases of matter, this research will explore several benchmark platforms where configurational entropy is utilized to tune magnetic, topological and correlated electron phases to establish new entropy-function relationships hitherto unrealized in a vast landscape of materials systems. In parallel, this project will have a major impact on the training of undergraduates, graduate students, and post-doctoral associates in synthesis and characterization of materials at the cutting edge of science and technology, and foster interaction of researchers from several different science and engineering disciplines at our universities and national laboratory institutions, and collaborative partnerships around the world. Bringing in both high-throughput synthesis and characterization, solid state chemistry and the computational framework of the AFLOW repository, the activities of this project will integrate experiments, computation and data-driven methods with interdisciplinary collaboration to establish the HEQM landscape as a potential new frontier for the Materials Genome Initiative. Technical description: This research addresses key questions in the new field of HEQMs, establishing a framework for a) exploring how entropy stabilizes new phases, b) identifying theoretical and experimental benchmarks, c) distinguishing high-entropy from non-high-entropy variants, and d) elucidating the role of entropy in emergent low-temperature properties. Key activities include discovery and exploration of new HEQMs using computational prediction and assisted learning techniques, and applying configurational entropy to known quantum materials systems. The focus on four example materials families amenable to high entropy configurations – spinel oxides, transition metal dichalcogenides, half-Heuslers and ThCr2Si2-type systems – will provide a template for benchmarking future directions of this approach. In parallel, establishing a dedicated repository for HEQM data will ensure that knowledge relating to synthesis activities is readily curated, stored and made accessible for analysis and mining, and generated materials data is prepared for public accessibility and application. Together, these activities will accelerate the development of HEQMs as a new avenue for discovery and optimization of quantum phenomena and applications. 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

Model theory studies the ways in which mathematical objects can be defined in some restricted formal language, and what structural properties are implied by these definability assumptions. It provides methods of converting asymptotic questions about finite structures into qualitative questions about the shape, volume or dimension of certain limiting infinite objects. This method of study originated in questions on foundations of mathematics, but in recent years it has found important applications in the study of some central objects of classical mathematics and computer science. The project investigates further these connections, with the major motivation of extending the existing techniques from binary structures (graphs) to structures of higher arity (hypergraphs), which represent a mathematical way of describing more complex networks in which interactions happen not just between two nodes at a time, but between multiple nodes simultaneously. This study will both deepen and extend the scope for applications of the infinitary model-theoretic machinery to questions in combinatorics of geometrically or algebraically arising hypergraphs, and conversely for applications of combinatorics to open questions in model theory. The project will involve training of graduate and undergraduate students. Shelah's classification program isolates combinatorial dividing lines (stability, distality, NIP, etc.) separating mathematical structures exhibiting various degrees of Gödelian behavior, from the tame ones in which one develops a “geometric” theory akin to algebraic geometry for definable sets in such structures. These tameness notions in Shelah’s classification theory are typically given by restrictions on the combinatorial complexity of definable binary relations. Many of the central results in graph combinatorics can be then improved dramatically if one restricts to graphs on the tame side of this classification, in particular to graphs arising from various algebraic or geometric configurations. The PI will investigate a higher generalization of Shelah's classification theory, where the restriction is only put on higher arity relations, focusing on n-dependence (with the case n=1 corresponding to the well studied class of NIP structures), n-stability, n-distality, and n-amalgamation, as opposed to the traditional binary case n=1. This will be applied to questions in extremal combinatorics of hypergraphs definable in various tame structures (via Keisler measures), as well as to generalizations of the polynomial expansion phenomena (Elekes-Szabó type theorems), and to the study of algebraic structures such as groups and fields definable in n-tame theories. 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

Synthetic biology is an emerging field that uses computation and design technology to modify biological cells for biological research and health applications. This new field is an important area for community involvement because synthetic biology has complex socioscientific implications and health applications. This project aims to engage youth and their families in this new field by developing participatory hands-on activities that involve living and biological materials as a first-hand context for engagement, synthetic biology will become accessible to these learners. Examples of potential biomaker activities include: an Agar Art activity (genetically modifying microbes to express color pigments that are then used to paint living art); Building with Mycelium activity (constructing and growing a substrate containing living fungus mycelium into a usable solid 3D object), and an Algae BioString activity (harvesting biological materials to design and create their own custom biostring). Participants will include middle school-aged youth and their families in Baltimore, Maryland, and rural areas in and around El Paso, Texas. Along with community partners and project staff, an initial group of youth and their families from each locale will collaboratively design and construct hands-on biomaker activities. These biomaker activities will be implemented through partner community organizations, reaching an estimated 500 youth and family participants. Resources designed for practitioners and researchers will include guidelines, lessons learned, and sample activities to help build educator knowledge and capacity to create and customize similar STEM experiences. Project findings and educational resources will be available on the project website and will be disseminated via conferences, research and education journals, and relevant networks such as REVISE (the Advancing Informal STEM Learning Program resource network). Project research is designed to advance knowledge on designing hands-on, engaging activities that involve living biological materials used in informal learning contexts. The project will research the co-design processes along with the impacts of the synthetic biology activities and intergenerational interactions within the communities involved. This project places primary emphasis on community contexts, using a participatory approach to address the following research questions: (1) how co-designing hands-on activities can align with local priorities, (2) how co-designed biomaker activities support personal understanding and attitude formation, and (3) how intergenerational dialogues shape synthetic biology knowledge, perceptions, and engagement. Both qualitative and quantitative data will be collected and analyzed to address these research questions, centered on hands-on, sustained, and engaging activities; STEM knowledge, attitudes, and interests; and intergenerational social processes for STEM learning. Primary data sources include transcripts of interviews and co-design session recordings. Secondary data sources include co-design and activity session artifacts and researcher observation notes. This Integrating Research and Practice project is funded by the Advancing Informal STEM Learning (AISL) program, which seeks to advance new approaches to, and evidence-based understanding of, the design and development of STEM learning in informal environments. This includes providing multiple pathways for engagement in STEM learning experiences. 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 project explores how information behaves in large-scale data storage systems that are designed to recover from hardware failures. Modern storage devices, such as glass-ceramic platters, arrange data in flat, grid-like patterns. To protect against data loss, information is encoded in a way that allows missing pieces to be reconstructed from surrounding bits, using a pre-set recovery rule. Over time, these systems handle vast and varied patterns of data. The research aims to understand how such data patterns evolve and organize themselves, from a statistical perspective. By identifying typical patterns and behaviors, the project will help improve the efficiency, reliability, and adaptability of storage systems—ultimately supporting better data preservation and easier access. In technical terms, the storage medium is modeled as an assignment of bits to nodes of an infinite two-dimensional grid, where the content of each node is determined by its neighbors according to a fixed recovery rule. A collection of such assignments—called a recoverable system—consists of all configurations satisfying this rule and forms the central object of study. The project aims to identify which recovery rules support systems that store the largest volume of the data on the medium, formalized as the topological entropy of the system. Introducing a temperature parameter into the model further increases in data density, assuming the system can tolerate a specified probability of errors. The system is defined through interaction energies between neighboring sites, with recoverable configurations corresponding to ground states. A key goal is to analyze equilibrium (Gibbs) distributions on the configuration space and to identify temperature regimes where phase transitions occur, leading to qualitatively different patterns of data stability. As another goal, the project aims at quantifying the system’s ability to retain information as it repeatedly restores data using the recovery rule over time, modeled as a Markov process on the space of the configurations. 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 I-Corps project is based on the development of a framework to answer complex spatial questions by using artificial intelligence (AI) to interpret a question and search for the answer. As AI becomes more prevalent, searches have become more interactive, flexible, and personalized through natural language interfaces and chatbots powered by large language models. Popular search engines provide an AI generated answer to many questions, even if they contain ambiguities or are under-specified. This response flexibility has not yet been extended to spatial searches, where questions tend to involve the names of points of interest or street addresses. To improve spatial searches, the solution develops a framework that handles complex spatial questions using AI to interpret the query and perform the search. Instead of searching by a name, address, or point of interest type, the search framework allows a user to provide a vague description of a place as a query, such as a "restaurant within a kilometer of a popular sporting venue or next to a bus stop." Searching such a query in any of the popular search engines or mapping platforms currently available yields an unsatisfactory result related to one of the points of interest in the query, without satisfying the spatial component. To correctly answer such a question, a person would need to visually search the appropriate areas on the map, which is time-consuming and tedious. A solution to this problem would lead to efficiencies across domains, such as military and defense, navigation, travel, and robotics sectors. This I-Corps project utilizes experiential learning coupled with a first-hand investigation of the industry ecosystem to assess the translation potential of the technology. This solution is based on the development of spatial pattern search techniques and language models to interpret spatial constraints from vague text. Using artificial intelligence to handle the natural language interpretation and infer details that are not specified directly in the text, the solution allows for a response that is spatially accurate and directly answers even vague questions, without incurring the computational burden of performing a poorly constrained spatial query using existing methods. By leveraging this technology, the project supports a new form of spatial search that has potential benefits in a variety of industries including military and defense, navigation, travel, and robotics. 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 Earth’s magnetic field is critical for sustaining our planet’s habitability, deflecting harmful solar radiation. However, the processes generating this field deep within the Earth’s core remain mysterious, hindering our ability to predict its future behavior – a concern given recent observations of relatively rapid changes in the field. This project addresses this fundamental challenge by developing a “digital twin” system for a large-scale laboratory experiment that mimics conditions inside the Earth. The digital twin will allow researchers to operate and optimize complex experiments remotely, explore scenarios inaccessible through physical experimentation alone, and ultimately improve our understanding of the forces shaping the geodynamo – the engine driving Earth’s magnetic field. By creating extensible tools for laboratory science, this research advances computational mathematics, supports training for a new generation of scientists and engineers, and has potential benefits for diverse fields including medical device design and external aerodynamics. This translational science collaborative project between University of Maryland (UM) and University of Illinois creates a digital twin consisting of the 3-meter liquid sodium geodynamo experiment at UM, coupled with advanced numerical modeling schemes based on high-order spectral element methods (Nek5000/RS) and data assimilation techniques including Ensemble Kalman Filters. The research team will develop Reduced Order Models ROM incorporating Deep Learning Neural Networks to enhance predictive capabilities and enable flow control strategies. By synchronizing the model with experimental observations, researchers aim to achieve a 1:1 correspondence between geometry changes and simulation results, ultimately allowing for bidirectional interaction between the physical experiment and its digital counterpart. This work will leverage high-performance computing resources to advance computational mathematics and provide a framework extensible to other laboratory science domains. 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

Computational linguistics – such as creating technology that can learn human language – has been central to the rise of artificial intelligence (AI). This conference grant brings together researchers from linguistics and computer science who study computational linguistics from different perspectives. Bringing these communities together enhances the nation’s AI workforce by broadening researchers’ expertise and, in the long term, providing a path towards developing AI that learns human language more efficiently. This is beneficial because it can reduce the costs involved in training and using AI, and because it supports the development of AI that understands a wider variety of languages. Having AI that learns like humans do could also provide a new tool for studying how humans learn language, leading to improved interventions for language disorders and more effective teaching of second languages in adulthood. Other benefits to society include educational opportunities for trainees that support workforce development for the AI and other language technology sectors. The conference grant supports the co-location of a premier conference on computational models of human language with a premier conference on language technology. This is the first time that the two conferences have been co-located. The conference features keynote speakers that do cutting-edge research at the intersection of both communities and a panel that directly addresses opportunities for integrating the two research areas. Conference scholarships enable researchers to attend both conferences, which allows researchers to engage in an extended period of critical interdisciplinary education and engagement. 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 I-Corps project focuses on the commercial potential of a new acoustic sensing solution that equips wearable audio devices with spatial hearing capabilities. The technology determines both the content of surrounding sounds and the direction they come from, enabling features such as distinguishing speakers in meetings, issuing directional safety alerts, and enhancing voice-based interactions in complex environments. The problem addressed is the absence of spatial awareness in current wearable technologies, which limits their effectiveness in dynamic, real-world settings. This limitation stems from the inability to integrate traditional multi-microphone systems into small form factors like earbuds. The solution responds to the growing societal demand for intelligent and intuitive human-technology interfaces across healthcare, mobility, workplace, and accessibility domains. By enhancing situational awareness in everyday wearables, the technology supports public safety and improves the functionality of voice-first systems for users. The technology advances the national health, prosperity, and welfare through access to intelligent tools, next-generation computing platforms, and the integration of advances in acoustics, materials, and machine learning. This I-Corps project utilizes experiential learning coupled with a first-hand investigation of the industry ecosystem to assess the translation potential of the technology. This solution is based on the development of a microstructure-assisted, acoustic front-end that captures directional information using a single microphone. The system encodes spatial cues into compact signal representations that are processed by a low-power neural network optimized for wearable devices. These spatial features are then aligned with speech embeddings and used as input to a language understanding model that operates in a cloud-supported architecture. The use of directional encoding with real-time, on-device processing minimizes latency and power consumption, making it suitable for continuous use in daily environments. By enabling wearable devices to interpret both linguistic content and physical sound location, the technology improves user interaction, safety, and privacy. The project explores user requirements, market fit, and technical constraints through direct engagements with stakeholders, ultimately guiding the transition of this innovation from research to practical deployment. 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

Saltwater intrusion is an often-invisible process that is challenging to identify until it has already caused substantial harm to coastal lands. Salty waters seep inland – above and below ground – salinizing soils and waters, devastating crop harvests, and burning forests from the inside out. In the low-lying Mid-Atlantic region, large areas of coastal farmland and forest have converted to marsh, causing substantial economic losses and damage to ecosystems. To address these pressing challenges, an assembled coalition of farmers, landowners, researchers, government, non-profits, and the private sector will work together to develop, evaluate, and implement science-based solutions, focused on two important coastal economic sectors: farming and forestry. By developing and implementing a portfolio of practical solutions, such as novel agricultural easements, web applications to map saltwater intrusion, market development for salt-tolerant crops, and alternative timber harvest strategies, the project will improve the resilience and well-being of rural coastal communities impacted by saltwater intrusion, now and in the future. Thus, the project will translate research into practical solutions to promote regional resilience through community-engaged team science. The project goal is to improve regional resilience across rural coastal lands affected by saltwater intrusion by extending the life of farms and forest tracts, reducing storm surge damage and revenue losses, and supporting regional terrestrial and aquatic biodiversity. We will achieve this by developing and implementing coordinated, community-engaged solutions, focusing on agricultural and forest lands in Maryland, Delaware and New Jersey. To co-develop solutions, the project will bring together leaders from academia, government (local to federal), non-profits, and the private sector—who often have worked to face these challenges in isolation. Building on recent advances in earth system science at the land-sea interface, knowledge of the region’s complex hydrological, ecological, geomorphological, biogeochemical, and human systems will be synthesized to develop and evaluate a portfolio of social, technological, and nature-based resilience strategies. Selection of solutions will be informed by both research and community input and assessed for feasibility, risk, cost, and benefit through approaches such as techno-economic analysis. A Guide to Coastal Resilience will be developed that details the coalition’s shared vision of resilience and coordinated implementation solutions and will be disseminated broadly to guide policy, investment, and advocacy. This coordinated effort will bridge the gap between basic science and practical, community-aligned resilience strategies to meet the region’s evolving challenges. 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 research project seeks to develop a rigorous theoretical foundation for amortized inference, a recent and impactful paradigmatic development in machine learning, statistics, and simulation. Amortization enables efficient, real-time responses to statistical queries by learning a model-dependent mapping from data to distributions, avoiding the need for expensive computations every time new data are presented. This capability underpins modern advances in generative AI, including diffusion models and variational autoencoders, with applications also extending to scientific machine learning (SciML) and simulation-based decision-making in operations research. Despite its widespread empirical success, fundamental questions persist: When do these methods work well, and when might they fail? How robust are the mappings to properties of the underlying problem? What kinds of statistical guarantees can be made about learned mappings, embodied for instance by deep neural networks? The goals of this project are twofold: (1) to deepen our understanding of the mathematical principles that underpin amortized inference, and (2) to inform the design of improved methods with provable guarantees. The project comprises three interrelated thrusts: 1) Functional Guarantees: This thrust investigates foundational properties of mappings from data to distributions: Do they exist? Are they unique? How well can they be approximated by, for example, neural networks? These results will elucidate the stability and robustness of amortized inference under data and model perturbations, 2) Statistical Guarantees: Building on the first thrust, this will establish both large-sample and finite-sample statistical guarantees for the learned mappings. The analysis will draw on techniques from M-estimation, approximation theory and Bayesian posterior contraction theory, 3) Methodological Developments: Existing amortized inference methods largely assume independent and identically distributed (i.i.d.) data. However, many applications-e.g., those involving data generated by Markov processes-violate this assumption. This thrust will extend amortized inference to non-i.i.d. settings, and will develop novel methodologies to fill this gap in the literature. Put together, these efforts aim to lay the theoretical groundwork for amortized inference, offering both insight and innovation in how statistical inference is carried out at scale. 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 project aims to develop and analyze tools from applied harmonic analysis with the goal of understanding and capitalizing on the formal concept of redundancy in mathematics and applications. The mathematical challenges involved in this project are formulated to directly contribute to more interpretable and explainable machine learning algorithms as well as address complex optimization problems that are currently intractable with existing computational resources. In addition to advancing foundational research, the project will play a key role in training graduate students, equipping them with cutting-edge skills in mathematics and computation. This will help cultivate a globally competitive STEM workforce capable of solving real-world challenges. The project will pursue two main research thrusts. The first thrust involves the development of both theory and algorithms for invariant coorbit representations. This includes designing stable Euclidean embeddings of the metric quotient space and analyzing their analytic and geometric properties. These embeddings will be applied to both optimization and machine learning tasks, with a focus on sorting-based neural architectures that yield easier interpretability and explainability. The second thrust involves the development and analyses of optimal factorizations of positive semi-definite self-adjoint operators and associated quadratic bounds. These results will be applied to a special form of the blind source separation problem. To achieve these goals, the research will integrate methods from optimization, harmonic analysis, computational invariant theory, and machine learning. This interdisciplinary approach aims to advance a mathematical framework for constructing invariant coorbit representations, deriving optimal factorizations of positive semi-definite operators, and computing quadratic bounds relevant to learning and optimization tasks. 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

Computing concepts are often highly abstract and can be difficult for introductory students to learn. Instructors use analogies to make these computing concepts more relatable and memorable. However, analogies may contain references that are not always relatable to students. This gap between instructor-student shared references and students' experiences can lead to students' confusion. Students can make their own analogies to learn course content, however, student analogy creation is largely an unguided process with that leads to mixed results. This project is designed to create classroom activities, tools, and an online library that will guide students as they create more robust and personalized analogies. This has the potential to help students more meaningfully and effectively learn computing concepts. By developing these shared learning tools, this project aims to improve comprehension, confidence, and persistence for learners from all backgrounds in introductory computing courses. As a result, this project has the potential to increase the nation’s pool of well-prepared computing professionals. This project is designed to address two primary challenges: (i) scaffolding novices to decompose abstract introductory computing concepts into key components, and (ii) guiding students to map key conceptual components onto personally meaningful analogous experiential domains. By developing digital scaffolding tools and peer- and instructor-feedback mechanisms that deliver timely formative assessment, the project aims to help learners test and refine their analogical conceptions. This could lead students to develop conceptual models that are more robust and less vulnerable to common naïve conceptions. The team will iteratively refine scaffolding and structured peer-feedback processes developed in past pilot studies. This work will be deployed in approximately eight introductory computing courses at multiple institutions. A mixed-methods, design-based framework will be used to guide iterative development of the scaffolding and feedback protocols. A quasi-experimental study will measure effects on concept-inventory scores, self-efficacy, engagement, and retention. This work will result in open-access templates, data sets, and a public repository of vetted student analogies, which will provide an evidence-based model for constructivist, personalized computing instruction. This project is funded by the STEM Education Postdoctoral Research Fellowship (STEM Ed PRF) Program that aims to enhance the research knowledge, skills, and practices of recent doctorates in STEM, STEM education, education, and related disciplines to advance their preparation to engage in fundamental and applied research that advances knowledge within the field. 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 objective of this Smart & Connected Communities - Development Grants (SCC-DG) project is to support research on co-design of bike networks that are safe, connected, and aligned with local needs. Cities struggle to design bikeways that balance technical constraints, safety considerations, lived realities, and aspirations of their citizens. This project explores how advances in generative artificial intelligence (AI) bridge the gap by integrating community input directly into the design process. During the planning phase, the team builds strong partnerships with key stakeholders, assesses how generative AI can support human-centered design, and refines key research questions on balancing engineering feasibility with community values. Infrastructure planning is driven by complex datasets of community needs, safety requirements, engineering constraints, and available funding. Current tools and methods often lack the ability to process varied information in an integrated way. The central research question guiding the planning effort is whether a generative AI-based system can effectively integrate multimodal and heterogeneous data inputs to produce bikeway designs that are technically feasible, regulation-compliant, and aligned with community priorities. Ultimately, this project contributes to advancing the responsible use of AI in participatory infrastructure design and planning. 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

Fire investigation training programs aim to equip investigators with the skills to identify fire origins and causes, but the chaotic nature of post-fire scenes presents substantial challenges. Investigators must connect evidence and scene features to the fire dynamics that shaped a scene, which requires strong spatial-temporal reasoning skills. Immersive training in realistic environments is essential to help investigators piece together evidence, analyze fire progression, and accurately trace fire origins. However, most training programs in the U.S. rely on lectures and 2D visuals, lacking the immersive experience needed to develop these crucial reasoning skills. Further, many investigators lack formal education in fire science, which is essential for understanding fire behavior. This project seeks to create a multimodal embodied training platform that advances fire investigation training through adaptive deliberate practice and learning analytics, focusing on the spatial-temporal reasoning skills needed to reconstruct fire development from observed fire damage and scene features. This new training approach will improve the quality and effectiveness of fire investigation practices, benefiting public safety by enabling more accurate identification of fire origins and causes. Many of the ideas can be extended to related fields such as crime scene investigation and other STEM areas requiring advanced spatial-temporal reasoning skills. To achieve these goals, the training platform will incorporate an AI-driven, physics-informed 3D fire modeling system that dynamically generates and visualizes fire scenarios based on learner-selected fire origins. Learners will identify and analyze scattered evidence, reconstruct fire progression, test interpretations, and explore variations in fire dynamics relative to observed damage patterns. Multimodal sensors will track learner interactions, enabling adaptive instructional approaches, enhancing engagement, and fostering seamless interactions between learners, instructors, and virtual fire scenarios. A deliberate practice pedagogical model will integrate structured skill-building exercises, multimodal analytics for performance assessment, and personalized adaptive training tailored to individual learner profiles. The platform's effectiveness will be evaluated in three phases: iterative expert reviews, student prototype assessments, and nationwide testing by early-career fire investigators, ensuring robust skill development in spatial-temporal reasoning for fire investigation. This project is funded by the Research on Innovative Technologies for Enhanced Learning (RITEL) program that supports early-stage exploratory research in emerging technologies for teaching and learning. 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 project will develop mathematical models that will aid in the understanding of animal migration. Migration is a widespread phenomenon that occurs seasonally as animals shift their locations in response to changing conditions. Oftentimes these changes involve spatial variation in resources that serve as cues for animals to track, resulting in wave-like population expansions. This research will use a series of novel mathematical modeling approaches to explore such seasonal, wave-like migratory dynamics, with a specific focus on understanding how the quality and quantity of resources interact to shape the pace and pattern of migration for varied theoretical scenarios. In addition, a pre-existing dataset of GPS tracking data for the critically endangered scimitar-horned oryx (Oryx dammah) will be analyzed to characterize when, where, and how well the animals track seasonal changes in resource availability in a resource-poor landscape. The project will support the training of undergraduate and graduate students who are developing skills and knowledge at the interface of mathematics and biology. Consumer tracking of transient resources occurs worldwide in a wide range of systems and taxa. The 'green wave surfing' hypothesis is a recent conceptual advance in understanding such resource tracking that is now widely discussed with regard to seasonal migrations of ungulates, birds, and marine species. According to this hypothesis, migrating consumer species living in seasonal systems should closely track the progression of the highly nutritious plant green-up wave that moves across the landscape as the growing season begins. Empirical data demonstrates that such tracking does occur for some individuals, populations, and species; however, 'surfing the green wave' is not universal, and instead some taxa either jump ahead of the green wave or lag behind it as it seasonally translates in space. The project will develop hybrid dynamical system models involving reaction-advection-diffusion equations with reaction and diffusion coefficients and growth governed by the quantity and quality of the resource green-up wave. Model variants including Allee effects, shifting habitats, and population structure will bring added biological realism. Research will address the impacts of sex- and age-specific migratory behaviors, predation, and mating success on migratory dynamics. Methods from differential equations, integral equations, and dynamical systems will be employed to identify conditions under which populations can persist in the long run. Existence of equilibrium solutions, traveling wave solutions, and oscillating solutions in time and density will be established to understand how 'surfing the green wave' promotes population growth and develops spatiotemporal patterns in population persistence on bounded domains. 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.