Key Takeaways
- Purdue University is the sole Indiana institution leading projects under the U.S. Department of Energy’s Genesis Mission awards, a highly competitive program that selected only 278 of 5,000 proposals nationwide.
- Two Purdue‑led initiatives focus on applying artificial intelligence (AI) to accelerate scientific discovery: one for real‑time data processing at the upcoming Electron‑Ion Collider (EIC) and another for improving the traceability of large language model (LLM) predictions.
- The EIC project, led by physicists Wei Xie and Miaoyuan Liu and computer scientist Ruqi Zhang, will develop autonomous, on‑sensor machine‑learning agents that compress data streams and highlight significant particle interactions.
- David F. Gleich’s project, conducted with Oak Ridge National Laboratory, investigates whether LLMs can directly cite their training data, a capability crucial for trustworthy scientific modeling.
- Five additional Purdue faculty contribute as collaborators on Genesis Mission awards headed by other institutions, spanning topics from critical‑mineral workflows to AI‑assisted code generation and nuclear‑reactor digital twins.
- These efforts align with Purdue’s Institute of Physical Artificial Intelligence and the broader Purdue Computes initiative, which emphasizes computing, physical AI, quantum science, and semiconductor innovation.
- Purdue’s recognition reflects its growing national stature in AI‑driven research, as highlighted by Executive Vice President for Research Dan DeLaurentis, who called the awards a “strong statement about where our faculty stand nationally.”
Purdue’s Dual Leadership in DOE Genesis Mission Awards
Purdue University stands out as the only university in Indiana to lead projects under the U.S. Department of Energy’s Genesis Mission awards, the DOE’s largest science and technology funding effort. Out of roughly 5,000 submissions from all 50 states, only 278 were selected, underscoring the competitiveness of the program. The awards target transformative research in energy, discovery science, and national security, with a particular emphasis on harnessing artificial intelligence to speed up scientific breakthroughs. Purdue’s success in securing two of these grants highlights the institution’s growing influence in AI‑focused research and its ability to attract federal investment for high‑impact, interdisciplinary work.
Executive Vice President Dan DeLaurentis on the Significance of the Awards
Dan DeLaurentis, Purdue’s executive vice president for research, praised the selections as a testament to the university’s national standing. He stated, “The selection of two Purdue projects for the Genesis Mission awards underscores our academic progress,” adding that “Successfully using AI to push the rate of scientific discovery can establish a new future in scientific discovery.” His remarks reflect the university’s strategic vision to integrate AI across disciplines, positioning Purdue as a leader in the next generation of data‑driven science.
Project One: Real‑Time AI for the Electron‑Ion Collider
The first Purdue‑led project, headed by Wei Xie (professor of physics and astronomy), Miaoyuan Liu (assistant professor of physics and astronomy), and Ruqi Zhang (assistant professor of computer science), concentrates on the Electron‑Ion Collider (EIC), a flagship U.S. research facility currently under construction. The team aims to design and develop autonomous, on‑sensor, real‑time machine‑learning models assisted by agentic AI. These models will identify important particle interactions, compress incoming data streams, and enable scientific instruments to discern which data are most valuable, thereby minimizing data loss. As Xie explained, “We will design autonomous, on-sensor and real-time machine learning models with agentic AI assistance intended to identify important particle interactions and compress data.”
Collaborative Partners for the EIC Initiative
To bring this ambitious vision to fruition, the Purdue team is partnering with Los Alamos National Laboratory and Fermi National Accelerator Laboratory. These collaborations combine expertise in high‑energy physics, detector design, and advanced computing, ensuring that the AI tools developed are both cutting‑edge and practical for the EIC’s unprecedented data rates. The partnership exemplifies the DOE’s goal of leveraging national laboratory resources alongside university innovation to tackle complex scientific challenges.
Project Two: Enhancing Traceability of Large Language Models
The second Purdue‑led effort is led by David F. Gleich, a professor of computer science, who will investigate the extent to which large language models (LLMs) can directly cite their training data. This capability is vital for scientific models because it allows researchers to understand the basis of AI‑generated predictions, fostering transparency and trust. Gleich’s work will be conducted in partnership with Oak Ridge National Laboratory, a leader in computational science and AI research. By probing LLMs’ ability to reference their sources, the project seeks to mitigate the “black‑box” nature of modern AI and improve reproducibility in scientific workflows.
Additional Purdue Contributions to Genesis Mission Awards
Beyond the two lead projects, five other Purdue faculty members serve as key collaborators on Genesis Mission awards headed by other institutions. David Nolte, the Edward M. Purcell Distinguished Professor of Physics and Astronomy, is contributing to an AI workflow for critical minerals in heterogeneous geosystems. Laura J. Pyrak‑Nolte, Distinguished Professor of Physics and Astronomy, is involved in two projects: one applying neural operators for rapid multiphysics simulation of enhanced geothermal systems, and another developing a physics‑informed multimodal model for fracture evolution in the upper crust. Xipeng Shen, professor of electrical and computer engineering, is working on an LLM workflow for code generation and optimization on emerging AI accelerators. Joerg Appenzeller, the Barry M. and Patricia L. Epstein Professor of Electrical and Computer Engineering, is researching Polaris‑Angstrom‑node equivalent SRAM and logic with polar materials. Finally, Stylianos Chatzidakis, nuclear engineering assistant professor and associate PUR‑1 director, is leading an AI‑driven workflow with cyber‑assured digital twin for small modular reactors (SMRs) and microreactors.
Integration with Purdue’s Institute of Physical Artificial Intelligence and Purdue Computes
These initiatives are anchored within Purdue’s Institute of Physical Artificial Intelligence, a core component of the broader Purdue Computes initiative. Purdue Computes emphasizes four pillars—computing departments, physical AI, quantum science, and semiconductor innovation—creating an ecosystem where AI research can intersect with cutting‑edge hardware and theoretical advances. The alignment ensures that the AI tools developed for the Genesis Mission projects benefit from, and contribute to, Purdue’s strengths in quantum computing, advanced materials, and high‑performance computing.
Purdue’s Institutional Context and Commitment to Accessibility
Purdue University, a top‑10 ranked public research institution in the United States, serves more than 106,000 students across multiple campuses, including over 57,000 at its main sites in West Lafayette and Indianapolis. As a land‑grant university committed to affordability, Purdue has frozen tuition for 14 consecutive years, enabling a growing number of students to graduate debt‑free. This commitment to accessibility complements its research excellence, ensuring that the breakthroughs generated by projects like the Genesis Mission awards can be educated and disseminated broadly.
Conclusion: A National Statement on AI‑Driven Discovery
The selection of two Purdue‑led projects for the DOE’s Genesis Mission awards, together with the participation of additional faculty in collaborative efforts, sends a strong signal about the university’s rising prominence in AI‑focused scientific research. As Dan DeLaurentis noted, the awards represent “a strong statement about where our faculty stand nationally.” By advancing real‑time data processing for the Electron‑Ion Collider and enhancing the traceability of large language models, Purdue is not only contributing to federal scientific priorities but also shaping the methodologies that will define the next era of discovery. Through interdisciplinary collaboration, strategic partnerships, and a commitment to accessible, high‑impact research, Purdue continues to cement its role as a national leader in the convergence of artificial intelligence and fundamental science.
https://www.purdue.edu/newsroom/2026/Q3/2-purdue-ai-research-projects-selected-among-the-first-genesis-mission-awards