Scientific outputs

Peer-reviewed articles, conference papers, book chapters, and reports spanning privacy-preserving machine learning, federated learning, differential privacy, and AI security and safety.

TOTAL PUBLICATIONS: 84  |  Google Scholar

Selected recent work at top-tier venues. The complete record follows below.

Grouped by year, reverse chronological. White papers and book chapters appear in dedicated sections below.

Preprints Under Review
Ordered by arXiv posting date, most recent first.

• Kotevska, O., Goethert, I., McGee, M., Mahbub, M., Wilkinson, S. R., Yip, R., Selvan, M. E., Gümüš, Z. H., Henschke, C., Klein, R. J., Morales, P., Aguayo, S. M., Danciu, I., & Shekar, M. C. (2026). Extending the Horizon of Early Diagnosis: Lung Cancer Prediction with Vision Transformers. arXiv preprint, submitted August 2026.
• Babu, A., Kaur, R., Pramanik, V., Kotevska, O., Bastian, N. D., Jha, S., Raj, S., Wu, Y., Jha, S. K., & Roy, A. (2026). Consilience: Conformally Calibrated Communication Control for Hidden-Profile Multi-Agent Reasoning. arXiv preprint, submitted August 2026.
• Riya, F. F., Hoque, S., Sun, J. S., & Kotevska, O. (2025). Accuracy is Not Enough: Poisoning Interpretability in Federated Learning via Color Skew. arXiv preprint, submitted November 2025.
• Mahbub, M., Klein, R. J., Selvan, M. E., Yip, R., Henschke, C., Morales, P., Goethert, I., Kotevska, O., Shekar, M. C., Wilkinson, S. R., McAllister, E., Aguayo, S. M., Gümüš, Z. H., Danciu, I., & VA Million Veteran Program. (2025). HEMERA: A Human-Explainable Transformer Model for Estimating Lung Cancer Risk using GWAS Data. arXiv preprint, submitted October 2025.

2026

• Pramanik, V., Maliha, M., Kotevska, O., Bastian, N. D., Jha, S., & Jha, S. K. (2026). Harmless in Pieces, Harmful in Motion: Detecting Multi-Agent Jailbreaks. NeurIPS 2026 — Conference on Neural Information Processing Systems.
• Pramanik, V., Maliha, M., Kotevska, O., Ramanathan, A., Bastian, N. D., Jha, S., & Jha, S. K. (2026). Unified Resource-Grounded Coordination Protocol for Orchestrator-Free Heterogeneous Multi-Agent Systems. NeurIPS 2026 — Conference on Neural Information Processing Systems.
• Xu, J., Hu, R., Kotevska, O., & Zhang, Z. (2026). Majority Bit-Aware Watermarking for Large Language Models. NeurIPS 2026 — Conference on Neural Information Processing Systems.
• Kotevska, O., Florey, D., Goldenberg, S., Isenberg, N. M., Raab, R., Schram, M., & Hodas, N. O. (2026). When Local Safety Does Not Compose: Measuring the Limits of Automated Risk Assessment for Cross-Institutional Agentic Systems. Trustworthy AI for Good Workshop at NeurIPS 2026.
• Zhang, Z., Hu, R., Kotevska, O., & Xu, J. (2026). SelfGrader: LLM Jailbreak Detection via Anchored Token-Level Logits. AACL-IJCNLP 2026 — Asia-Pacific Chapter of the Association for Computational Linguistics and International Joint Conference on Natural Language Processing, Findings.
• Kumar, R., Kotevska, O., & Polyzou, A. (2026). Forgetting Only What Matters: Layer-Selective Unlearning toward Robust LLMs. AACL-IJCNLP 2026 — Asia-Pacific Chapter of the Association for Computational Linguistics and International Joint Conference on Natural Language Processing.
• Tran, T., Kotevska, O., & Xiong, L. (2026). Automated Membership Inference Attacks: Discovering MIA Signal Computations using LLM Agents. Transactions on Machine Learning Research (TMLR).
• Kotevska, O., Carlson, M., Gao, Y., Jeanson, F., Li, Y., Lindskog, W., Naseri, M., Ryu, M., Tyagi, S., Watkins, J., Wang, F., Madduri, R., & Kim, K. (2026). From Pilots to Production: Lessons in Cross-Institutional Federated Training and AI for Science. SC26 — 4th TPC Workshop: Building Open AI Infrastructure, Models, and Agentic Systems for Science.
• Kotevska, O., Bastian, N. D., Jha, S., Ferreira da Silva, R., Bellet, A., Madduri, R., Hu, R., & Kim, K. (2026). Privacy Foundations for Multi-Institutional Scientific Artificial Intelligence. S-HPC'26 — 5th Annual Workshop on Cyber Security in High-Performance Computing, co-located with SC26.
• Zhang, Z., Hu, R., Kotevska, O., & Xu, J. (2026). Vision Token Manipulation Attacks on Cloud-Edge Inference of Large Vision-Language Models. GLOBECOM 2026 — IEEE Global Communications Conference, Communication & Information System Security.
• Riya, F. F., Kotevska, O., & Sun, J. S. (2026). IntraShuffler: A Privacy Preserving Framework for Heterogeneous DP Federated Learning. IFIP Annual Conference on Data and Applications Security and Privacy, 40–61. DOI
• Xu, J., Hu, R., Kotevska, O., & Zhang, Z. (2026). XMark: Reliable Multi-Bit Watermarking for LLM-Generated Texts. ACL 2026 — 64th Annual Meeting of the Association for Computational Linguistics, 14747–14763. DOI Code
• Xu, J., Hu, R., Kotevska, O., & Zhang, Z. (2026). Traceable Black-box Watermarks for Federated Learning. ICLR 2026 — International Conference on Learning Representations. Code
• Pramanik, V., Maliha, M., Jha, S., Velasquez, A., Kotevska, O., & Jha, S. K. (2026). Selective Amnesia using Contrastive Subnet Erasure for Class Level Unlearning in Vision Models. CVPR 2026 — IEEE/CVF Conference on Computer Vision and Pattern Recognition. Highlight Code
• Babu, A., Kaur, R., Bastian, N., Kotevska, O., Jha, S., Wu, Y., Jha, S., & Roy, A. (2026). CTRL-STEER: Closed-Loop Neuron Activation Control in Vision-Language-Action Models. Visual Concepts Workshop at CVPR 2026.
• Kotevska, O., Nguyen, T., Ferreira da Silva, R., Engelmann, C., & Balaprakash, P. (2026). Scalable Federated Learning for Scientific Foundation Models on Leadership-Class Systems. EuroMLSys @ EuroSys 2026, 439–446. DOI
• Kotevska, O., Nguyen, T., & Hernandez, O. (2026). Energy–Performance Trade-offs in Federated Learning with SmartNIC-Enabled Communication on HPC Systems. IPDPSW 2026 — Accelerators and Hybrid Emerging Systems Workshop.
• Kotevska, O., Patton, R., Jha, S., & Balaprakash, P. (2026). DP-TwoLevel: Two-Stage Gradient Subspace Learning for Differentially Private Federated Learning. SPIE Conference on Assurance and Security for AI-enabled Systems, 14. DOI
• Pramanik, V., Kotevska, O., Bastian, N., & Jha, S. (2026). Entropy-Weighted Multi-Layer Attention for Token-Level Attribution in Autoregressive Language Models. SPIE Conference on Assurance and Security for AI-enabled Systems, 48. DOI
• Xu, J., Hu, R., & Kotevska, O. (2026). Optimal Client Sampling in Federated Learning with Client-level Heterogeneous Differential Privacy. IEEE Internet of Things Journal. DOI Code
• Pramanik, V., Kotevska, O., Velasquez, A., Jha, A., & Jha, K. S. (2026). SPUN: Spectral Projection-based UNlearning in Hyperdimensional Computing. AAAI Workshop on Artificial Intelligence for Cyber Security.

2025

• Kotevska, O., Yang, W., & Al-Masri, E. (2025). Engineering Privacy at the Edge: A Practical Guide to Differential Privacy in System Architectures. IEEE ICCD 2025 — 43rd International Conference on Computer Design, 872–875. DOI Code
• Tyagi, S., Cozma, A., Kotevska, O., & Wang, F. (2025). OmniFed: A Modular Framework for Configurable Federated Learning from Edge to HPC. ACM/IEEE SC Workshop on Extreme Heterogeneity and AI Convergence in HPC, 516–523. DOI Code
• Yang, W., Al-Masri, E., & Kotevska, O. (2025). MIC-DP: A Scalable Correlation-Aware Differential Privacy Framework for Edge AI and High-Dimensional Data. IEEE Transactions on Privacy. DOI Code
• Kotevska, O., He, X., & Al-Masri, E. (2025). Enhancing Smart Home Privacy: A Tutorial on Local Differential Privacy Techniques for Frequency and Mean Estimation. IEEE Communications Magazine, 63(8), 116–122. DOI
• Kotevska, O. (2025). Privacy Preservation from High-Performance Computing to Autonomous Science. IEEE Computational Intelligence Magazine, 20(3), 5–6. DOI
• Riya, F. F., Hoque, S., Yang, Y., Sun, J., & Kotevska, O. (2025). Balancing Trade-offs: Adaptive Differential Privacy in Interpretable Machine Learning Models. IEEE PST 2025 — Conference on Privacy, Security, and Trust, 1–6. DOI
• Afrose, S., & Kotevska, O. (2025). Dynamical Sketching for Enhanced Communication Efficiency in Federated Learning. IEEE International Conference on Artificial Intelligence, 982–987. DOI

2024

• Kim, K., Raghavan, K., Kotevska, O., Dorier, M., Madduri, R., Ryu, M., et al., & Yousefian, F. (2024). Privacy-Preserving Federated Learning for Science: Challenges and Research Directions. IEEE BigData 2024 — IEEE International Conference on Big Data, 7849–7853. DOI
• Shu, Y., Kotevska, O., Reshniak, V., & Sedovak, A. (2024). Assessing Membership Inference Attacks under Distribution Shifts. IEEE BigData 2024 — IEEE International Conference on Big Data, 4127–4131. DOI
• Kotevska, O. (2024). Privacy by Design in Distributed Edge Systems: Innovating Secure Workflows for Smart Cities. IEEE Smart Cities Newsletter.
• Brogan, J., Kotevska, O., Torres, A., Jha, S., & Adams, M. (2024). Improving Robustness of Spectrogram Classifiers with Neural Stochastic Differential Equations. IEEE MLSP 2024 — 34th International Workshop on Machine Learning for Signal Processing, 1–6. DOI
• Zhang, Y., Zhao, Y., Li, Z., Cheng, X., Wang, Y., Kotevska, O., Yu, P. S., & Derr, T. (2024). A Survey on Privacy in Graph Neural Networks: Attacks, Preservation, and Applications. IEEE Transactions on Knowledge and Data Engineering, 36(12), 7497–7515. Repository DOI
• Sances, R., Kotevska, O., & Laiu, P. (2024). Frequency Oracle for Sensitive Data Monitoring. Proceedings of the AAAI Conference on Artificial Intelligence, 38(21), 23642–23643. DOI

2023

• Kotevska, O., & Kusne, A. G. (2023). Privacy-Preserving Federated Learning with Frequency Estimation. NeurIPS 2023 Workshop.
• Al-Masri, E., Souri, A., Mohamed, H., Yang, W., Olmsted, J., & Kotevska, O. (2023). Energy-Efficient Cooperative Resource Allocation and Task Scheduling for Internet of Things Environments. Elsevier Internet of Things, 23, 100832. DOI

2022

• Mohamed, H., Al-Masri, E., Kotevska, O., & Souri, A. (2022). A Multi-Objective Approach for Optimizing Edge-based Resource Allocation using TOPSIS. MDPI Electronics, 11(18), 2888. DOI
• Joy, D., Kotevska, O., & Al-Masri, E. (2022). Investigating Users Privacy Concerns of Internet of Things (IoT) Smart Devices. IEEE ECICE 2022, 70–76. DOI
• Yu, P. S., Kotevska, O., & Derr, T. (2022). PAS: Privacy Algorithms in Systems. ACM CIKM 2022, 5181–5182. DOI
• Kotevska, O., Johnson, J., & Kusne, A. G. (2022). Analyzing Data Privacy for Edge Systems. IEEE SMARTCOMP 2022, 223–228. DOI
• Tombs, V., Kotevska, O., & Young, S. (2022). Privacy Amplification for Episodic Training Methods. ACM CIKM Workshop on Privacy Algorithms in Systems.
• Kusne, A. G., & Kotevska, O. (2022). Privacy Policy Robustness to Reverse Engineering. ACM CIKM Workshop on Privacy Algorithms in Systems.

2021

• Afrose, S., Yao, D. D., & Kotevska, O. (2021). Measurement of Local Differential Privacy Techniques for IoT-based Streaming Data. IEEE PST 2021 — Conference on Privacy, Security, and Trust, 1–10. DOI
• Kotevska, O., Alamudun, F., & Stanley, C. (2021). Optimal Balance of Privacy and Utility with Differential Privacy Deep Learning Frameworks. IEEE CSCI 2021, 425–430. DOI
• Peralta-Peterson, M., & Kotevska, O. (2021). Effectiveness of Privacy Techniques in Smart Metering Systems. IEEE CSCI 2021, 675–678. DOI
• Kotevska, O., Munk, J., Kurte, K., Du, Y., Amasyali, K., Smith, R. W., & Zandi, H. (2021). Methodology for interpretable reinforcement learning model for HVAC energy control. IEEE BigData 2021, 1555–1564. ★ Best Paper Runner-Up DOI
• Amasyali, K., Kurte, K., Zandi, H., Munk, J., Kotevska, O., & Smith, R. (2021). Double Deep Q-Networks for Optimizing Electricity Cost of a Water Heater. IEEE PES ISGT 2021, 1–5. DOI
• Gao, S., Kotevska, O., Sorokine, A., & Christian, J. B. (2021). A pre-training and self-training approach for biomedical named entity recognition. PLOS One, 16(2), e0246310. DOI
• Herrmannova, D., Thakur, G., Grant, J. N., Tansakul, V., Eaton, B., Kotevska, O., Burdette, J., Smyth, M., & Smith, M. (2021). Challenges in Automated Detection of COVID-19 Misinformation. ACM Workshop on Human Aspects of Misinformation Online.
• Du, Y., Li, F., Munk, J., Kurte, K., Kotevska, O., Amasyali, K., & Zandi, H. (2021). Multi-task deep reinforcement learning for intelligent multi-zone residential HVAC control. Elsevier Electric Power Systems Research, 192, 106959. DOI
• Du, Y., Zandi, H., Kotevska, O., Kurte, K., Munk, J., & Li, F. (2021). Intelligent multi-zone residential HVAC control strategy based on deep reinforcement learning. Elsevier Applied Energy, 281, 116117. ★ 2022 Highly Cited Research Paper Award DOI

2020

• Kurte, K., Munk, J., Amasyali, K., Kotevska, O., Cui, B., Kuruganti, T., & Zandi, H. (2020). Electricity Pricing Aware Deep Reinforcement Learning based Intelligent HVAC Control. ACM SenSys Workshop on Reinforcement Learning for Energy Management in Buildings & Cities, 6–10. DOI
• Thakur, G., & Kotevska, O. (2020). Activity Characterization for Modeling Behavioral-Driven Human Mobility in Platial Networks. Proceedings of the 4th ACM SIGSPATIAL Workshop on Location-Based Recommendations, Geosocial Networks, and Geoadvertising, 1–5. DOI
• Perumalla, K., Lopez, J., Alam, M., Kotevska, O., Hempel, M., & Sharif, H. (2020). A Novel Vetting Approach to Cybersecurity Verification in Energy Grid Systems. IEEE KPEC 2020, 1–6. DOI
• Kotevska, O., Kurte, K., Munk, J., Johnston, T., McKee, E., Perumalla, K., & Zandi, H. (2020). RL-HEMS: Reinforcement Learning-based Home Energy Management System for HVAC Energy Optimization. ASHRAE Transactions, 126, 421–429.
• McKee, E., Du, Y., Li, F., Munk, J., Johnston, T., Kurte, K., et al., & Zandi, H. (2020). Deep reinforcement learning for residential HVAC control with consideration of human occupancy. IEEE PESGM 2020, 1–5. DOI
• Kurte, K., Munk, J., Kotevska, O., Amasyali, K., Smith, R., McKee, E., Du, Y., Cui, B., Kuruganti, T., & Zandi, H. (2020). Evaluating the Adaptability of Reinforcement Learning based HVAC Control for Residential Homes. MDPI Sustainability, 12(18), 7727. DOI

2019 and before

• Kotevska, O., Perumalla, K., & Lopez, J. (2019). Kensor: Coordinated Intelligence from Co-Located Sensors. IEEE BigData 2019, 4169–4174. DOI
• Kotevska, O., Kusne, A. G., Samarov, D. V., Lbath, A., & Battou, A. (2017). Dynamic Network Model for Smart City Data-Loss Resilience Case Study: City-to-City Network for Crime Analytics. IEEE Access, 5, 20524–20535. DOI
• Kotevska, O., & Lbath, A. (2017). Sentiment analysis of Social Sensors for improvement of local services. International Journal of Computing and Digital Systems, 6(4).
• Kotevska, O., Lbath, A., & Bouzefrane, S. (2016). Toward a Real-Time Framework in Cloudlet-Based Architecture. Tsinghua Science and Technology Journal, 21(1), 80–88. DOI
• Kotevska, O., Gelernter, J., & Lbath, A. (2016). Event model to facilitate data sharing among services: Closing the gap between research and implementation. IEEE WF-IoT 2016, 577–584. DOI
• Kotevska, O., Padi, S., & Lbath, A. (2016). Automatic Categorization of Social Sensor Data. Elsevier Procedia Computer Science, 98, 596–603. DOI
• Kotevska, O., Lbath, A., & Bouzefrane, S. (2015). Toward a Framework in Cloudlet-Based Architecture for a Real-Time Predictional Model. IEEE UIC-ATC-ScalCom 2015, 1016–1021. DOI
• Koceski, S., Kotevska, O., Vlahu-Gjorgievska, E., & Trajkovic, V. (2014). Continuous real-time monitoring of patient's vital signs based on ZigBee standard. International Journal of Informatics and Communication Technology (IJ-ICT), 3(2), 73. DOI
• Kotevska, O., Vlahu-Gjorgievska, E., & Koceski, S. (2013). Using ZigBee low-power wireless standard for monitoring patients' signs. Elsevier ICT Innovation.
• Gjorgievska, E. V., Kotevska, O., & Trajkovik, V. (2012). Collaboration Model within Personal Healthcare System. ISGT 2012.
• Kotevska, O., Vlahu-Gjorgievska, E., Trajkovic, V., & Koceski, S. (2012). Towards a patient-centered collaborative health care system model. International Journal of Computer Theory and Engineering (IJCTE), 1025–1029. DOI
• Kotevska, O., Vlahu-Gjorgievska, E., & Trajkovik, V. (2011). COHESY - Collaborative Health Care System Model. CIIT 2011.

• Rath, S., Abdeen, Z., Kotevska, O., Reshniak, V., & Singh, V. (2026). The Security of Reinforcement Learning Systems in Electric Grid Domain. IEEE-Wiley: AI for Cybersecurity: Research and Practice, 505–532. DOI
• Brogan, J., Passarella, L., Adams, M., Phathanapirom, B., Martindale, N., Kotevska, O., Yohe, M., Kerekes, R., & Stewart, S. (2026). Robust AI Techniques to Support High Consequence Applications in the Cyber Age. IEEE-Wiley: AI for Cybersecurity: Research and Practice.
• Kotevska, O., & Andelfinger, P. (2022). Reinforcement Learning for Intelligent Building Energy Management System Control. Intelligent Data Mining and Analysis in Power and Energy Systems — Wiley, 371–386. DOI
• (2021). An Open Call to Solve Scientific Data Challenges Using Advanced Data Analytics and Edge Computing. Springer — Driving Scientific and Engineering Discoveries Through the Integration of Experiment, Big Data, and Modeling and Simulation. DOI
• (2020). Data Challenge 2020: An Open Call to Solve Data Problems in the Areas of Neutron Science, Material Science, Urban Modeling and Dynamics, Geophysics, and Biomedical Informatics. Springer — Driving Scientific and Engineering Discoveries Through the Convergence of HPC, Big Data and AI. DOI

Responses to federal solicitations, agency RFIs, and technical reports. These represent policy influence distinct from the peer-reviewed record.