Searcharxiv⌕ Search

arXiv subjects

Maanak Gupta

Publications and source records attributed to Maanak Gupta.

At least 19 recordsLinked to original sources
arXiv2026

ExplainGuard: A Zero Trust Framework for Post-Hoc Explanation Integrity Guarantees in Blackbox XAI Models

As machine learning (ML) models are increasingly deployed in high-stakes environments, explainable AI (XAI) methods like SHAP and LIME have become essential for regulatory compliance and trust. However, the current auditing paradigm relies on an implicit "chain of trust" where third-party auditors are assumed to be trusted. Recent research demonstrates that this assumption is flawed and adversarial auditors can manipulate XAI explanations through manipulation attacks such as output shuffling or scaffolding out-of-distribution (OOD) to conceal model biases while maintaining high prediction accuracy aiming for fairwashed explanation. In this paper, we introduce a novel defense framework, ExplainGuard, that leverages a Zero-Trust architecture (ZTA) design to be incorporated within the XAI explanation supply chain and ensures the integrity of the generated explanation. This framework would help us to replace the ambiguous default assumption of "auditor is trustworthy," with a continuous "verify-then-trust" approach. Our design architecture establishes a Policy Decision Point (PDP) that enforces three distinct pillars of verification before any explanation is released to the user: (1) asset integrity via behavioral fingerprint to detect model substitution, (2) semantic validity using axiomatic consistency checks to reject mathematically impossible explanations, and (3) feature faithfulness verification utilizing a ranking stability approach with minimal computational overhead. Finally, we evaluate how ExplainGuard can effectively neutralize state- of-the-art explanation manipulation attacks while transforming the auditing process into a verifiable operation.

cs.CR↗
arXiv2026

Malformer: A Multi-Modal Malware Detector Using Transformers

Traditional malware detection systems that rely on a single representation of malware often fail to identify novel threats. These representations of malware binaries, also known as modalities, do not provide the models with sufficient information to discriminate among all samples. Additionally, individual representations introduce new failure modes, with some modality extraction being dependent upon the success of disassembling. Past works have integrated either additional modalities or more discriminative representations for classification. In this work, we present Malformer, a quadrimodal malware detection model that incorporates text, image, graph, and audio representations of Windows executables. We demonstrate that multimodal transformer fusion can enhance the performance of Windows malware detectors over that of unimodal and bimodal detectors. Malformer employs a combination of two RoBERTa encoders paired with a modified Vision Transformer for image data, WavLM for audio data, and an adaptive loss-weighting scheme to fuse modality-specific representations. Evaluated on a dataset of 201,549 binary samples, Malformer achieved 98.3% accuracy and an F1 score of 0.9833, outperforming both unimodal baselines and bimodal detectors by 4.6-17.6 percentage points. Malformer demonstrates that multimodal fusion provides a promising foundation for countering the growing scale of malware threats, equipping defenders with generalized and resilient detection capabilities.

cs.CR↗
arXiv2026

PANOPTICON: A PII-Based Assemblage of Naturalistic Output Tokens for Investigating Privacy Leakage Within LLM Context Window

Large Language Models (LLMs) are capable of generalizing human language for the completion of never-before-seen tasks, leading to widespread deployment. While this automation provides clear utility, completing these tasks often requires the insertion of Personally Identifiable Information (PII), strings of information that uniquely identify some individual, raising privacy concerns. However, ethics has prevented the curation of a public, authentic dataset of PII. Without an appropriate dataset, it is difficult to quantify privacy risks. Thus, we introduce the PANOPTICON pipeline and dataset. The dataset, generated by Meta's Llama-3.1-8B-Instruct model, contains 67, 718 prompts, intended for the models context window, containing PII spans derived from 9,674 publicly available synthetic user profiles. We measure lexical diversity and S-BERT diversity of the created dataset to evaluate realism. Finally, we present a case study showcasing the utility of PANOPTICON data for understanding Prompt Inversion Attacks (PIAs). PANOPTICON thus emerges as the first benchmark dataset for studying PIAs over private corpora, providing a foundation for future LLM privacy research.

cs.AI↗
arXiv2026

VEXAIoT: Autonomous IoT Vulnerability EXploitation using AI Agents

Internet of Things (IoT) systems are inherently vulnerable due to constrained hardware, outdated firmware, and insecure default configurations, creating a need for scalable and adaptive security testing approaches. While recent adoptions of Large Language Model (LLM) agents have demonstrated promise in penetration testing and Capture-the-Flag (CTF) environments, their application to IoT specific vulnerabilities remains unexplored. This paper presents an autonomous multi-agent framework, referred to as Vulnerability EXploitation using AI Agents (VEXAIoT), for vulnerability discovery and exploitation in IoT environments using LLM-based reasoning and offensive security tools. The framework combines a vulnerability detection agent and an attack execution agent to perform reconnaissance, plan attack sequences, and execute exploits against vulnerable IoT services. The system is evaluated in IoTGoat and Metasploitable environments across ten attack scenarios mapped to OWASP IoT vulnerabilities. Experimental results show attack success rate of up to 100% with low token overhead and average execution times under two minutes for most attacks. Across 260 attack executions, VEXAIoT achieves a 95.0% overall success rate, including 94.5% success in IoTGoat and 96.7% success in Metasploitable2. These results demonstrate the potential for LLM-driven agents to automate IoT vulnerability assessment and offensive security workflows in controlled environments

cs.CR↗
arXiv2026

Knowledge Base Poisoning Attacks and Defense for Policy-Aware LLM-RAG Framework

This paper presents an adversarial security study of the Policy-Aware LLM Retrieval-Augmented Generation (PA-LLM-RAG) framework for Internet of Battlefield Things (IoBT) mission control. We propose Query-Agnostic Semantic Retrieval Poisoning, a novel attack that injects semantically crafted rules into the IoBT knowledge base achieving high retrieval ranking across all operator query types without requiring knowledge of runtime prompts. The attack achieves 85% LLM context corruption from a single injected rule (1.6% poisoning rate) and saturates at 7.7% poisoning, demonstrating that even minimal knowledge base compromise is sufficient to corrupt mission decisions. To counter this threat, we propose CLD-KB (Cyber-Layered Defense for Knowledge Base), a dual-detector anomaly detection framework combining One-Class SVM boundary detection with a novel Member-Based Category Spread analysis that exploits the three-category IoBT policy taxonomy to identify poisoned rules before they reach the decision LLM. CLD-KB significantly outperforms five baseline methods including DBSCAN, LOF, K-Means, Isolation Forest, and One-Class SVM in both poisoning detection and knowledge preservation, with only 7ms computational overhead per mission, establishing it as an effective and edge-deployable defense for LLM-driven IoBT mission systems.

cs.CR↗
arXiv2026

Policy-Aware Edge LLM-RAG Framework for Internet of Battlefield Things Mission Orchestration

Large Language Models (LLMs) offer a promising interface for intent-driven control of autonomous cyber-physical systems, but their direct use in mission-critical Internet of Battlefield Things (IoBT) environments raises significant safety, reliability, and policy-compliance concerns. This paper presents a Policy-Aware Large Language Model Retrieval-Augmented Generation (referred as PA-LLM-RAG), an edge-deployed LLM orchestration framework for IoBT mission control that integrates retrieval-augmented reasoning and independent command verification. The proposed PA-LLM-RAG framework combines a lightweight retrieval module that grounds decisions in operational policies and telemetry with a locally hosted LLM for mission planning and a secondary JudgeLLM for validating user generated commands prior to execution. To evaluate PA-LLM-RAG, we implement a simulated IoBT environment using RoboDK and assess four open-source LLMs across controlled mission scenarios of increasing complexity, including baseline operations, threat detection, coverage recovery, multi-event coordination, and policy-violation requests. Experimental results demonstrate that the framework effectively detects policy-violating commands while maintaining low-latency response suitable for edge deployment. Gemma-2B achieving the highest overall reliability with 4.17 sec latency and 100% success rate. The findings highlight a clear tradeoff between reasoning capacity and responsiveness across models and show that combining deterministic safeguards with JudgeLLM verification significantly improves reliability in LLM-driven IoBT orchestration.

cs.NI↗
arXiv2026

Explainability-Guided Adversarial Attacks on Transformer-Based Malware Detectors Using Control Flow Graphs

Transformer-based malware detection systems operating on graph modalities such as control flow graphs (CFGs) achieve strong performance by modeling structural relationships in program behavior. However, their robustness to adversarial evasion attacks remains underexplored. This paper examines the vulnerability of a RoBERTa-based malware detector that linearizes CFGs into sequences of function calls, a design choice that enables transformer modeling but may introduce token-level sensitivities and ordering artifacts exploitable by adversaries. By evaluating evasion strategies within this graph-to-sequence framework, we provide insight into the practical robustness of transformer-based malware detectors beyond aggregate detection accuracy. This paper proposes a white-box adversarial evasion attack that leverages explainability mechanisms to identify and perturb most influential graph components. Using token- and word-level attributions derived from integrated gradients, the attack iteratively replaces positively attributed function calls with synthetic external imports, producing adversarial CFG representations without altering overall program structure. Experimental evaluation on small- and large-scale Windows Portable Executable (PE) datasets demonstrates that the proposed method can reliably induce misclassification, even against models trained to high accuracy. Our results highlight that explainability tools, while valuable for interpretability, can also expose critical attack surfaces in transformer-based malware detectors.

cs.CR↗
arXiv2026

Privacy Enhanced PEFT: Tensor Train Decomposition Improves Privacy Utility Tradeoffs under DP-SGD

Fine-tuning large language models on sensitive data poses significant privacy risks, as membership inference attacks can reveal whether individual records were used during training. While Differential Privacy (DP) provides formal protection, applying DP to conventional Parameter-Efficient Fine-Tuning (PEFT) methods such as Low-Rank Adaptation (LoRA) often incurs substantial utility loss. In this work, we show that a more structurally constrained PEFT architecture, Tensor Train Low-Rank Adaptation (TTLoRA), can improve the privacy-utility tradeoff by shrinking the effective parameter space while preserving expressivity. To this end, we develop TTLoRA-DP, a differentially private training framework for TTLoRA. Specifically, we extend the ghost clipping algorithm to Tensor Train cores via cached contraction states, enabling efficient Differentially Private Stochastic Gradient Descent (DP-SGD) with exact per-example gradient norm computation without materializing full per-example gradients. Experiments on GPT-2 fine-tuning over the Enron and Penn Treebank datasets show that TTLoRA-DP consistently strengthens privacy protection relative to LoRA-DP while maintaining comparable or better downstream utility. Moreover, TTLoRA exhibits lower membership leakage even without DP training, using substantially smaller adapters and requiring on average 7.6X fewer parameters than LoRA. Overall, our results demonstrate that TTLoRA offers a practical path to improving the privacy-utility tradeoff in parameter-efficient language model adaptation.

cs.CR↗
arXiv2026

A Survey of Agentic AI and Cybersecurity: Challenges, Opportunities and Use-case Prototypes

Agentic AI marks an important transition from single-step generative models to systems capable of reasoning, planning, acting, and adapting over long-lasting tasks. By integrating memory, tool use, and iterative decision cycles, these systems enable continuous, autonomous workflows in real-world environments. This survey examines the implications of agentic AI for cybersecurity. On the defensive side, agentic capabilities enable continuous monitoring, autonomous incident response, adaptive threat hunting, and fraud detection at scale. Conversely, the same properties amplify adversarial power by accelerating reconnaissance, exploitation, coordination, and social-engineering attacks. These dual-use dynamics expose fundamental gaps in existing governance, assurance, and accountability mechanisms, which were largely designed for non-autonomous and short-lived AI systems. To address these challenges, we survey emerging threat models, security frameworks, and evaluation pipelines tailored to agentic systems, and analyze systemic risks including agent collusion, cascading failures, oversight evasion, and memory poisoning. Finally, we present three representative use-case implementations that illustrate how agentic AI behaves in practical cybersecurity workflows, and how design choices shape reliability, safety, and operational effectiveness.

cs.CR↗
arXiv2025

RAG-targeted Adversarial Attack on LLM-based Threat Detection and Mitigation Framework

The rapid expansion of the Internet of Things (IoT) is reshaping communication and operational practices across industries, but it also broadens the attack surface and increases susceptibility to security breaches. Artificial Intelligence has become a valuable solution in securing IoT networks, with Large Language Models (LLMs) enabling automated attack behavior analysis and mitigation suggestion in Network Intrusion Detection Systems (NIDS). Despite advancements, the use of LLMs in such systems further expands the attack surface, putting entire networks at risk by introducing vulnerabilities such as prompt injection and data poisoning. In this work, we attack an LLM-based IoT attack analysis and mitigation framework to test its adversarial robustness. We construct an attack description dataset and use it in a targeted data poisoning attack that applies word-level, meaning-preserving perturbations to corrupt the Retrieval-Augmented Generation (RAG) knowledge base of the framework. We then compare pre-attack and post-attack mitigation responses from the target model, ChatGPT-5 Thinking, to measure the impact of the attack on model performance, using an established evaluation rubric designed for human experts and judge LLMs. Our results show that small perturbations degrade LLM performance by weakening the linkage between observed network traffic features and attack behavior, and by reducing the specificity and practicality of recommended mitigations for resource-constrained devices.

cs.CR↗
arXiv2025

LLM-based Multi-class Attack Analysis and Mitigation Framework in IoT/IIoT Networks

The Internet of Things has expanded rapidly, transforming communication and operations across industries but also increasing the attack surface and security breaches. Artificial Intelligence plays a key role in securing IoT, enabling attack detection, attack behavior analysis, and mitigation suggestion. Despite advancements, evaluations remain purely qualitative, and the lack of a standardized, objective benchmark for quantitatively measuring AI-based attack analysis and mitigation hinders consistent assessment of model effectiveness. In this work, we propose a hybrid framework combining Machine Learning (ML) for multi-class attack detection with Large Language Models (LLMs) for attack behavior analysis and mitigation suggestion. After benchmarking several ML and Deep Learning (DL) classifiers on the Edge-IIoTset and CICIoT2023 datasets, we applied structured role-play prompt engineering with Retrieval-Augmented Generation (RAG) to guide ChatGPT-o3 and DeepSeek-R1 in producing detailed, context-aware responses. We introduce novel evaluation metrics for quantitative assessment to guide us and an ensemble of judge LLMs, namely ChatGPT-4o, DeepSeek-V3, Mixtral 8x7B Instruct, Gemini 2.5 Flash, Meta Llama 4, TII Falcon H1 34B Instruct, xAI Grok 3, and Claude 4 Sonnet, to independently evaluate the responses. Results show that Random Forest has the best detection model, and ChatGPT-o3 outperformed DeepSeek-R1 in attack analysis and mitigation.

cs.CR↗
arXiv2025

Safety and Security Analysis of Large Language Models: Benchmarking Risk Profile and Harm Potential

While the widespread deployment of Large Language Models (LLMs) holds great potential for society, their vulnerabilities to adversarial manipulation and exploitation can pose serious safety, security, and ethical risks. As new threats continue to emerge, it becomes critically necessary to assess the landscape of LLMs' safety and security against evolving adversarial prompt techniques. To understand the behavior of LLMs, this research provides an empirical analysis and risk profile of nine prominent LLMs, Claude Opus 4, DeepSeek V3 (both open-source and online), Gemini 2.5 Flash, GPT-4o, Grok 3, Llama 4 Scout, Mistral 7B, and Qwen 3 1.7B, against 24 different security and safety categories. These LLMs are evaluated on their ability to produce harmful responses for adversarially crafted prompts (dataset has been made public) for a broad range of safety and security topics, such as promotion of violent criminal behavior, promotion of non-violent criminal activity, societal harms related to safety, illegal sexual content, dangerous code generation, and cybersecurity threats beyond code. Our study introduces the Risk Severity Index (RSI), an agile and scalable evaluation score, to quantify and compare the security posture and creating a risk profile of LLMs. As the LLM development landscape progresses, the RSI is intended to be a valuable metric for comparing the risks of LLMs across evolving threats. This research finds widespread vulnerabilities in the safety filters of the LLMs tested and highlights the urgent need for stronger alignment, responsible deployment practices, and model governance, particularly for open-access and rapidly iterated models.

cs.CR↗
arXiv2025

Bridging Cloud Convenience and Protocol Transparency: A Hybrid Architecture for Ethereum Node Operations on Amazon Managed Blockchain

As blockchain technologies are increasingly adopted in enterprise and research domains, the need for secure, scalable, and performance-transparent node infrastructure has become critical. While self-hosted Ethereum nodes offer operational control, they often lack elasticity and require complex maintenance. This paper presents a hybrid, service-oriented architecture for deploying and monitoring Ethereum full nodes using Amazon Managed Blockchain (AMB), integrated with EC2-based observability, IAM-enforced security policies, and reproducible automation via the AWS Cloud Development Kit. Our architecture supports end-to-end observability through custom EC2 scripts leveraging Web3.py and JSON-RPC, collecting over 1,000 real-time data points-including gas utilization, transaction inclusion latency, and mempool dynamics. These metrics are visualized and monitored through AWS CloudWatch, enabling service-level performance tracking and anomaly detection. This cloud-native framework restores low-level observability lost in managed environments while maintaining the operational simplicity of managed services. By bridging the simplicity of AMB with the transparency required for protocol research and enterprise monitoring, this work delivers one of the first reproducible, performance-instrumented Ethereum deployments on AMB. The proposed hybrid architecture enables secure, observable, and reproducible Ethereum node operations in cloud environments, suitable for both research and production use.

cs.CR↗
arXiv2025

TT-LoRA MoE: Unifying Parameter-Efficient Fine-Tuning and Sparse Mixture-of-Experts

We propose Tensor-Trained Low-Rank Adaptation Mixture of Experts (TT-LoRA MoE), a novel computational framework integrating Parameter-Efficient Fine-Tuning (PEFT) with sparse MoE routing to address scalability challenges in large model deployments. Unlike traditional MoE approaches, which face substantial computational overhead as expert counts grow, TT-LoRA MoE decomposes training into two distinct, optimized stages. First, we independently train lightweight, tensorized low-rank adapters (TT-LoRA experts), each specialized for specific tasks. Subsequently, these expert adapters remain frozen, eliminating inter-task interference and catastrophic forgetting in multi-task setting. A sparse MoE router, trained separately, dynamically leverages base model representations to select exactly one specialized adapter per input at inference time, automating expert selection without explicit task specification. Comprehensive experiments confirm our architecture retains the memory efficiency of low-rank adapters, seamlessly scales to large expert pools, and achieves robust task-level optimization. This structured decoupling significantly enhances computational efficiency and flexibility: uses only 2% of LoRA, 0.3% of Adapters and 0.03% of AdapterFusion parameters and outperforms AdapterFusion by 4 value in multi-tasking, enabling practical and scalable multi-task inference deployments.

cs.LG↗
arXiv2024

A Lightweight Edge-CNN-Transformer Model for Detecting Coordinated Cyber and Digital Twin Attacks in Cooperative Smart Farming

The agriculture sector is increasingly adopting innovative technologies to meet the growing food demands of the global population. To optimize resource utilization and minimize crop losses, farmers are joining cooperatives to share their data and resources among member farms. However, while farmers benefit from this data sharing and interconnection, it exposes them to cybersecurity threats and privacy concerns. A cyberattack on one farm can have widespread consequences, affecting the targeted farm as well as all member farms within a cooperative. In this research, we address existing gaps by proposing a novel and secure architecture for Cooperative Smart Farming (CSF). First, we highlight the role of edge-based DTs in enhancing the efficiency and resilience of agricultural operations. To validate this, we develop a test environment for CSF, implementing various cyberattacks on both the DTs and their physical counterparts using different attack vectors. We collect two smart farming network datasets to identify potential threats. After identifying these threats, we focus on preventing the transmission of malicious data from compromised farms to the central cloud server. To achieve this, we propose a CNN-Transformer-based network anomaly detection model, specifically designed for deployment at the edge. As a proof of concept, we implement this model and evaluate its performance by varying the number of encoder layers. Additionally, we apply Post-Quantization to compress the model and demonstrate the impact of compression on its performance in edge environments. Finally, we compare the model's performance with traditional machine learning approaches to assess its overall effectiveness.

cs.CR↗
arXiv2024

Explainable Artificial Intelligence (XAI) for Malware Analysis: A Survey of Techniques, Applications, and Open Challenges

Machine learning (ML) has rapidly advanced in recent years, revolutionizing fields such as finance, medicine, and cybersecurity. In malware detection, ML-based approaches have demonstrated high accuracy; however, their lack of transparency poses a significant challenge. Traditional black-box models often fail to provide interpretable justifications for their predictions, limiting their adoption in security-critical environments where understanding the reasoning behind a detection is essential for threat mitigation and response. Explainable AI (XAI) addresses this gap by enhancing model interpretability while maintaining strong detection capabilities. This survey presents a comprehensive review of state-of-the-art ML techniques for malware analysis, with a specific focus on explainability methods. We examine existing XAI frameworks, their application in malware classification and detection, and the challenges associated with making malware detection models more interpretable. Additionally, we explore recent advancements and highlight open research challenges in the field of explainable malware analysis. By providing a structured overview of XAI-driven malware detection approaches, this survey serves as a valuable resource for researchers and practitioners seeking to bridge the gap between ML performance and explainability in cybersecurity.

cs.CR↗
arXiv2024

SoK: Leveraging Transformers for Malware Analysis

The introduction of transformers has been an important breakthrough for AI research and application as transformers are the foundation behind Generative AI. A promising application domain for transformers is cybersecurity, in particular the malware domain analysis. The reason is the flexibility of the transformer models in handling long sequential features and understanding contextual relationships. However, as the use of transformers for malware analysis is still in the infancy stage, it is critical to evaluate, systematize, and contextualize existing literature to foster future research. This Systematization of Knowledge (SoK) paper aims to provide a comprehensive analysis of transformer-based approaches designed for malware analysis. Based on our systematic analysis of existing knowledge, we structure and propose taxonomies based on: (a) how different transformers are adapted, organized, and modified across various use cases; and (b) how diverse feature types and their representation capabilities are reflected. We also provide an inventory of datasets used to explore multiple research avenues in the use of transformers for malware analysis and discuss open challenges with future research directions. We believe that this SoK paper will assist the research community in gaining detailed insights from existing work and will serve as a foundational resource for implementing novel research using transformers for malware analysis.

cs.CR↗
arXiv2024

Explainability-Informed Targeted Malware Misclassification

In recent years, there has been a surge in malware attacks across critical infrastructures, requiring further research and development of appropriate response and remediation strategies in malware detection and classification. Several works have used machine learning models for malware classification into categories, and deep neural networks have shown promising results. However, these models have shown its vulnerabilities against intentionally crafted adversarial attacks, which yields misclassification of a malicious file. Our paper explores such adversarial vulnerabilities of neural network based malware classification system in the dynamic and online analysis environments. To evaluate our approach, we trained Feed Forward Neural Networks (FFNN) to classify malware categories based on features obtained from dynamic and online analysis environments. We use the state-of-the-art method, SHapley Additive exPlanations (SHAP), for the feature attribution for malware classification, to inform the adversarial attackers about the features with significant importance on classification decision. Using the explainability-informed features, we perform targeted misclassification adversarial white-box evasion attacks using the Fast Gradient Sign Method (FGSM) and Projected Gradient Descent (PGD) attacks against the trained classifier. Our results demonstrated high evasion rate for some instances of attacks, showing a clear vulnerability of a malware classifier for such attacks. We offer recommendations for a balanced approach and a benchmark for much-needed future research into evasion attacks against malware classifiers, and develop more robust and trustworthy solutions.

cs.CR↗