Publications

Preprints

1. Plale, B., Karthikeyan, N., Gamage, I., Stubbs, J., & Withana, S. AI/ML Model Cards in Edge AI Cyberinfrastructure: towards Agentic AI. arXiv:2511.21661.

   arXiv BIB Abstract

   @unpublished{plale2025modelcards,
     author = {Plale, Beth and Karthikeyan, Neelesh and Gamage, Isuru and Stubbs, Joe and Withana, Sachith},
     title = {AI/ML Model Cards in Edge AI Cyberinfrastructure: towards Agentic AI},
     note = {arXiv:2511.21661},
     doi = {2511.21661}
   }
   

   A framework for implementing AI/ML model cards within edge AI computing environments, with a focus on supporting agentic AI systems. Model cards serve as documentation artifacts that capture essential metadata about machine learning models, including their intended use, performance characteristics, and limitations. Standardized model card implementations become increasingly important in distributed edge computing contexts where models are deployed across heterogeneous infrastructures, facilitating transparency and reproducibility in agentic AI applications.

2. Plale, B., Jyesta, S. N., & Withana, S. Vector embedding of multi-modal texts: a tool for discovery? arXiv:2509.08216.

   arXiv BIB Abstract

   @unpublished{plale2025embedding,
     author = {Plale, Beth and Jyesta, Sai Navya and Withana, Sachith},
     title = {Vector embedding of multi-modal texts: a tool for discovery?},
     note = {arXiv:2509.08216},
     doi = {2509.08216}
   }
   

   Investigates whether vector embeddings of multi-modal documents can effectively support information discovery tasks, examining how embeddings that integrate both textual and visual content from documents perform in retrieval and discovery scenarios compared to traditional single-modality approaches.

Publications

1. Vallabhajosyula, M. S., Freeman, N., Karthikeyan, N., Padhy, S., Garcia, C. R., Molakalmuru, G. G., Khuvis, S., Stubbs, J., Jamthe, A., Ramnath, R., & Plale, B. (2026). Configuring, Training, and Deploying AI Applications Through Integrated Gateways and Frameworks for Scientific Field Research. SN Computer Science, 7.

   DOI BIB Abstract

   @article{vallabhajosyula2026gateways,
     title = {Configuring, Training, and Deploying AI Applications Through Integrated Gateways and Frameworks for Scientific Field Research},
     author = {Vallabhajosyula, Manikya Swathi and Freeman, Nathan and Karthikeyan, Neelesh and Padhy, Smruti and Garcia, Christian R. and Molakalmuru, Gautam Gururaj and Khuvis, Samuel and Stubbs, Joe and Jamthe, Anagha and Ramnath, Rajiv and Plale, Beth},
     journal = {SN Computer Science},
     volume = {7},
     year = {2026},
     doi = {10.1007/s42979-026-05092-4}
   }
   

   Presents a framework distributing AI activities across cyberinfrastructure – from resource-limited edge devices to centralized HPC/cloud environments. Edge systems handle model fine-tuning and real-time inference tasks, while centralized systems manage large-scale training and version control, outlining MLOps stages and demonstrating how middleware simplifies implementation for scientific field research.

2. Vallabhajosyula, M. S., Molakalmuru, G. G., Khuvis, S., Karthikeyan, N., Freeman, N., Garcia, C., Stubbs, J., Plale, B., & Ramnath, R. (2026). From HPC to Edge: A Web-Based Workflow for AI Model Testing and Deployment. Proceedings of the Practice and Experience in Advanced Research Computing 2026: Resilient Roots + Empowered Communities.

   DOI BIB Abstract

   @inproceedings{vallabhajosyula2026hpcedge,
     title = {From HPC to Edge: A Web-Based Workflow for AI Model Testing and Deployment},
     author = {Vallabhajosyula, Manikya Swathi and Molakalmuru, Gautam Gururaj and Khuvis, Samuel and Karthikeyan, Neelesh and Freeman, Nathan and Garcia, Christian and Stubbs, Joe and Plale, Beth and Ramnath, Rajiv},
     booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing 2026: Resilient Roots + Empowered Communities},
     year = {2026},
     doi = {10.1145/3785462.3815843}
   }
   

   Artificial intelligence (AI) workflows increasingly span heterogeneous environments, from centralized high-performance computing (HPC) systems to resource-constrained edge devices. A primary hurdle in these pipelines is the configuration mismatch between development and deployment, often requiring researchers to manually rewrite scripts for specific edge requirements. This poster demonstrates an integrated management framework – showcasing both web-based and standalone user interface (UI) elements – powered by Tapis for the systematic testing and deployment of machine learning models. Building on existing cyberinfrastructure including the ML Field Planner for configuration, PATRA for model management, and the Cyberinfrastructure Knowledge Network (CKN) for telemetry, the framework provides a unified control interface. By utilizing the ML Edge Server as a consistent, pluggable runtime across the edge-to-center continuum, the system ensures that configurations and telemetry translate directly to field execution. This demonstration shows how a unified UI-driven workflow streamlines the transition from initial evaluation to deployment, ensuring operational consistency and reproducibility without manual script porting.

3. Plale, B., Karthikeyan, N., Gamage, I., Stubbs, J., & Withana, S. (2025). AI/ML Model Cards in Edge AI Cyberinfrastructure: towards Agentic AI. 2025 IEEE International Conference on e-Science (e-Science).

   BIB Abstract

   @inproceedings{plale2025modelcards_ieee,
     title = {AI/ML Model Cards in Edge AI Cyberinfrastructure: towards Agentic AI},
     author = {Plale, Beth and Karthikeyan, Neelesh and Gamage, Isuru and Stubbs, Joe and Withana, Sachith},
     booktitle = {2025 IEEE International Conference on e-Science (e-Science)},
     year = {2025},
     organization = {IEEE}
   }
   

   Published proceedings version. A framework for implementing AI/ML model cards within edge AI computing environments, with a focus on supporting agentic AI systems and standardized documentation for transparency and reproducibility in agentic AI applications.

4. Stubbs, J., Balasubramaniam, S., Khuvis, S., Withana, S., Vallabhajosyula, M. S., Cardone, R., Garcia, C., Freeman, N., Guzman, C., Plale, B., Ramnath, R., & Berger-Wolf, T. (2025). ML Field Planner: Analyzing and Optimizing ML Pipelines For Field Research. PEARC ’25: Practice and Experience in Advanced Research Computing 2025: The Power of Collaboration.

   DOI BIB Abstract

   @inproceedings{stubbs2025mlfieldplanner,
     title = {ML Field Planner: Analyzing and Optimizing ML Pipelines For Field Research},
     author = {Stubbs, Joe and Balasubramaniam, Sowbaranika and Khuvis, Samuel and Withana, Sachith and Vallabhajosyula, Manikya Swathi and Cardone, Richard and Garcia, Christian and Freeman, Nathan and Guzman, Carlos and Plale, Beth and Ramnath, Rajiv and Berger-Wolf, Tanya},
     booktitle = {PEARC '25: Practice and Experience in Advanced Research Computing 2025: The Power of Collaboration},
     year = {2025},
     doi = {10.1145/3708035.3736013}
   }
   

   Best Paper in Applications and Software, and Phil Andrews Award for Best Paper Overall. Introduces ML Field Planner, a tool for analyzing and optimizing machine learning pipelines used in field research settings.

5. Withana, S., & Plale, B. (2024). Patra ModelCards: AI/ML Accountability in the Edge-Cloud Continuum. 2024 IEEE 20th International Conference on e-Science (e-Science).

   BIB Abstract

   @inproceedings{withana2024patramodelcards,
     title = {Patra ModelCards: AI/ML Accountability in the Edge-Cloud Continuum},
     author = {Withana, Sachith and Plale, Beth},
     booktitle = {2024 IEEE 20th International Conference on e-Science (e-Science)},
     year = {2024},
     organization = {IEEE}
   }
   

   Introduces a framework for Model Cards, Patra ModelCards, that embeds model cards in the edge-cloud continuum for semi-automated information capture with the objective of greater trustworthiness and accountability for AI/ML models, including fairness, explainability, and behavior of a model in different deployed environments.

6. Marru, S., Pierce, M., Plale, B., Pamidighantam, S., Wannipurage, D., Christie, M., Ranawaka, I., Abeysinghe, E., Quick, R., Tajkhorshid, E., Koric, S., Basney, J., Spivak, M., Isralewitz, B., Bernardi, R., Gomes, D., Krishnan, G., Bazhenov, M., Smallen, S., … Yoshimoto, K. (2023). Cybershuttle: An End-to-End Cyberinfrastructure Continuum to Accelerate Discovery in Science and Engineering. PEARC ’23: Practice and Experience in Advanced Research Computing 2023.

   DOI BIB Abstract

   @inproceedings{marru2023cybershuttle,
     title = {Cybershuttle: An End-to-End Cyberinfrastructure Continuum to Accelerate Discovery in Science and Engineering},
     author = {Marru, Suresh and Pierce, Marlon and Plale, Beth and Pamidighantam, Sudhakar and Wannipurage, Dimuthu and Christie, Marcus and Ranawaka, Isuru and Abeysinghe, Eroma and Quick, Rob and Tajkhorshid, Emad and Koric, Seid and Basney, Jim and Spivak, Mariano and Isralewitz, Barry and Bernardi, Rafael and Gomes, Diego and Krishnan, Giri and Bazhenov, Maxim and Smallen, Shava and Majumdar, Amit and Arkhipov, Anton and Dai, Kael and Liu, Xiao-Ping and Yoshimoto, Kenneth},
     booktitle = {PEARC '23: Practice and Experience in Advanced Research Computing 2023},
     year = {2023},
     doi = {10.1145/3569951.3593602}
   }
   

   Presents Cybershuttle, an end-to-end cyberinfrastructure continuum designed to accelerate discovery in science and engineering by connecting researchers’ desktops to advanced computing, data, and software resources.