Piyush Nagasubramaniam
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Piyush Nagasubramaniam has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Neeraj Karamchandani
Department of Computer Science and Engineering, The Pennsylvania State University, University Park, PA 16802, USA
4 shared publications - Sencun Zhu
Department of Computer Science and Engineering, The Pennsylvania State University, University Park, PA 16802, USA
4 shared publications - Chen Wu
The Pennsylvania State University
3 shared publications - Dinghao Wu
The Pennsylvania State University
1 shared publication - Xinhong Xie
Department of Computer Science and Engineering, The Pennsylvania State University, University Park, PA 16802, USA
1 shared publication - Yongzhong He
Beijing Key Laboratory of Security and Privacy in Intelligent Transportation, School of Computer, Beijing Jiaotong University, Beijing, China
1 shared publication
Latest Publications
- Open Access
LLM-Based Penetration Testing in the Presence of Honeypots
Large language model (LLM) agents are increasingly being employed for offensive cybersecurity tasks such as automated vulnerability discovery, reconnaissance, and penetration testing. This new capability also threatens one of the defender’s most valuable tools: deception. Traditional honeypots rely on realism and obscurity to lure human or script-driven attackers into revealing tactics, techniques, and procedures (TTPs), but LLM-driven attackers can reason about heterogeneous artifacts and use the honeypot suspicion to guide target-selection decisions. We study host-level inference-budget allocation for LLM penetration-testing agents facing honeypot uncertainty, using a heuristic policy that couples detector evidence, vulnerability severity, realized host-specific cost, and post-connect verification. We formalize the attacker’s problem as a budgeted decision process: an agent interacts with potential targets, consuming LLM execution budget during reconnaissance and exploitation, and must decide whether to engage (continue exploitation) or disengage (skip) when honeypot suspicion arises. In a controlled mixed-host testbed, we examine detector behavior, target ranking, post-entry expenditure, and genuine-host confirmations. In the evaluated testbed, the proposed policy achieves higher mean genuine-host confirmation counts than the baselines, with the LLM attacker configuration held constant across policies. While defenses are beyond our present scope, we discuss implications for future adversarially resilient and adaptive honeypot design.
Open AccessLLM-Based Penetration Testing in the Presence of Honeypots
Large language model (LLM) agents are increasingly being employed for offensive cybersecurity tasks such as automated vulnerability discovery, reconnaissance, and penetration testing. This new capability also threatens one of the defender’s most valuable tools: deception. Traditional honeypots rely on realism and obscurity to lure human or script-driven attackers into revealing tactics, techniques, and procedures (TTPs), but LLM-driven attackers can reason about heterogeneous artifacts and use the honeypot suspicion to guide target-selection decisions. We study host-level inference-budget allocation for LLM penetration-testing agents facing honeypot uncertainty, using a heuristic policy that couples detector evidence, vulnerability severity, realized host-specific cost, and post-connect verification. We formalize the attacker’s problem as a budgeted decision process: an agent interacts with potential targets, consuming LLM execution budget during reconnaissance and exploitation, and must decide whether to engage (continue exploitation) or disengage (skip) when honeypot suspicion arises. In a controlled mixed-host testbed, we examine detector behavior, target ranking, post-entry expenditure, and genuine-host confirmations. In the evaluated testbed, the proposed policy achieves higher mean genuine-host confirmation counts than the baselines, with the LLM attacker configuration held constant across policies. While defenses are beyond our present scope, we discuss implications for future adversarially resilient and adaptive honeypot design.
- Open Access
- 8 July 2026
- Published by MDPI AG in Journal of Cybersecurity and Privacy
- Vol. 6 (4) , 121
Text-to-Unlearn: Robust Concept Removal in GANs via Text Prompts
State-of-the-art generative models exhibit powerful image-generation capabilities, raising ethical and legal challenges for service providers. Consequently, Content Removal Techniques (CRTs) have emerged to control outputs without requiring full retraining. However, the problem of unlearning in Generative Adversarial Networks (GANs) remains largely unexplored. We propose Text-to-Unlearn, a novel framework that selectively unlearns concepts from pre-trained GANs using only text prompts, enabling feature and identity unlearning, as well as fine-grained tasks such as expression and multi-attribute removal in models trained on human faces. Our approach leverages natural language descriptions to guide unlearning without additional datasets or supervised finetuning, offering a scalable solution. To evaluate the effectiveness of our method, we introduce an automated unlearning assessment method using state-of-the-art image–text alignment metrics and propose a new metric: degree of unlearning. Additionally, we assess robustness by introducing adversarial attacks to subvert unlearning. Our results demonstrate that Text-to-Unlearn achieves robust unlearning, resisting adversarial attempts to recover erased concepts while preserving model utility. To our knowledge, this is the first cross-modal unlearning framework for GANs, advancing the management of generative model behavior.
- 8 July 2026
- Published by MDPI AG in Journal of Cybersecurity and Privacy
- Vol. 6 (4) , 121
Open AccessText-to-Unlearn: Robust Concept Removal in GANs via Text Prompts
State-of-the-art generative models exhibit powerful image-generation capabilities, raising ethical and legal challenges for service providers. Consequently, Content Removal Techniques (CRTs) have emerged to control outputs without requiring full retraining. However, the problem of unlearning in Generative Adversarial Networks (GANs) remains largely unexplored. We propose Text-to-Unlearn, a novel framework that selectively unlearns concepts from pre-trained GANs using only text prompts, enabling feature and identity unlearning, as well as fine-grained tasks such as expression and multi-attribute removal in models trained on human faces. Our approach leverages natural language descriptions to guide unlearning without additional datasets or supervised finetuning, offering a scalable solution. To evaluate the effectiveness of our method, we introduce an automated unlearning assessment method using state-of-the-art image–text alignment metrics and propose a new metric: degree of unlearning. Additionally, we assess robustness by introducing adversarial attacks to subvert unlearning. Our results demonstrate that Text-to-Unlearn achieves robust unlearning, resisting adversarial attempts to recover erased concepts while preserving model utility. To our knowledge, this is the first cross-modal unlearning framework for GANs, advancing the management of generative model behavior.
- 24 April 2026
- Published by Springer Nature
- p. 369-389
Antitoxin: A Framework for Controlling Persistent Backdoors in Federated Learning
- 24 April 2026
- Published by Springer Nature
- p. 369-389
Antitoxin: A Framework for Controlling Persistent Backdoors in Federated Learning
- Open Access
- 7 November 2024
- Published by MDPI AG in Applied Sciences
- Vol. 14 (22) , 10254
Privacy-Preserving Live Video Analytics for Drones via Edge Computing
The use of lightweight drones has surged in recent years across both personal and commercial applications, necessitating the ability to conduct live video analytics on drones with limited computational resources. While edge computing offers a solution to the throughput bottleneck, it also opens the door to potential privacy invasions by exposing sensitive visual data to risks. In this work, we present a lightweight, privacy-preserving framework designed for real-time video analytics. By integrating a novel split-model architecture tailored for distributed deep learning through edge computing, our approach strikes a balance between operational efficiency and privacy. We provide comprehensive evaluations on privacy, object detection, latency, bandwidth usage, and object-tracking performance for our proposed privacy-preserving model.
- 7 November 2024
- Published by MDPI AG in Applied Sciences
- Vol. 14 (22) , 10254
Open AccessPrivacy-Preserving Live Video Analytics for Drones via Edge Computing
The use of lightweight drones has surged in recent years across both personal and commercial applications, necessitating the ability to conduct live video analytics on drones with limited computational resources. While edge computing offers a solution to the throughput bottleneck, it also opens the door to potential privacy invasions by exposing sensitive visual data to risks. In this work, we present a lightweight, privacy-preserving framework designed for real-time video analytics. By integrating a novel split-model architecture tailored for distributed deep learning through edge computing, our approach strikes a balance between operational efficiency and privacy. We provide comprehensive evaluations on privacy, object detection, latency, bandwidth usage, and object-tracking performance for our proposed privacy-preserving model.