Mohan Gurusamy

Aditya Nawal, Manit Baser, Mohan Gurusamy

AI agents augment large language models with external tools such as web retrieval, enabling grounded and up-to-date responses. However, incorporating external content into the generation pipeline can weaken the safety alignment mechanisms that govern model outputs. Prior work shows that enabling retrieval in agents increases compliance with harmful requests. We introduce AgentREVEAL, a diagnostic framework for analyzing retrieval-induced safety degradation in LLM agents. The framework examines two axes: how retrieval is integrated into the agent pipeline and the properties of the retrieved content. Along the integration axis, we find that binding tool invocation and response generation in a single step amplifies harmful outputs. Along the content axis, we uncover the Safe Source Paradox: even oppositional or safety-oriented sources, such as pages containing warnings or risk disclaimers, can increase harmful compliance by an average of 25% compared to the no-retrieval baseline. Finally, we show that relevance acts as a shared activation condition for both vulnerabilities. Similar patterns appear on frontier closed models, and harmful compliance remains elevated under several representative pipeline interventions, with some agents also entering this regime under autonomous retrieval. Because relevance is also what makes retrieval useful, these results expose a safety-utility trade-off for retrieval-enabled agents. We introduce HarmURLBench, a benchmark containing 1,405 real-world URLs paired with 320 harmful behaviors to support future evaluations.

Poorvi Joshi, Alakesh Kalita, Mohan Gurusamy

Internet of Things (IoT) involves sensors for monitoring and wireless networks for efficient communication. However, resource-constrained IoT devices and limitations in existing wireless technologies hinder its full potential. Integrating Unmanned Aerial Vehicles (UAVs) into IoT networks can address some challenges by expanding its' coverage, providing security, and bringing computing closer to IoT devices. Nevertheless, effective data collection in UAV-assisted IoT networks is hampered by factors, including dynamic UAV behavior, environmental variables, connectivity instability, and security considerations. In this survey, we first explore UAV-based IoT networks, focusing on communication and networking aspects. Next, we cover various UAV-based data collection methods their advantages and disadvantages, followed by a discussion on performance metrics for data collection. As this article primarily emphasizes reliable and efficient data collection in UAV-assisted IoT networks, we briefly discuss existing research on data accuracy and consistency, network connectivity, and data security and privacy to provide insights into reliable data collection. Additionally, we discuss efficient data collection strategies in UAV-based IoT networks, covering trajectory and path planning, collision avoidance, sensor network clustering, data aggregation, UAV swarm formations, and artificial intelligence for optimization. We also present two use cases of UAVs as a service for enhancing data collection reliability and efficiency. Finally, we discuss future challenges in data collection for UAV-assisted IoT networks.

Binghui Wu, Philipp Gysel, Dinil Mon Divakaran et al.

Recent research works have proposed machine learning models for classifying IoT devices connected to a network. However, there is still a practical challenge of not having all devices (and hence their traffic) available during the training of a model. This essentially means, during the operational phase, we need to classify new devices not seen in the training phase. To address this challenge, we propose ZEST -- a ZSL (zero-shot learning) framework based on self-attention for classifying both seen and unseen devices. ZEST consists of i) a self-attention based network feature extractor, termed SANE, for extracting latent space representations of IoT traffic, ii) a generative model that trains a decoder using latent features to generate pseudo data, and iii) a supervised model that is trained on the generated pseudo data for classifying devices. We carry out extensive experiments on real IoT traffic data; our experiments demonstrate i) ZEST achieves significant improvement (in terms of accuracy) over the baselines; ii) SANE is able to better extract meaningful representations than LSTM which has been commonly used for modeling network traffic.

Anum Talpur, Mohan Gurusamy

Integration of machine learning (ML) in 5G-based Internet of Vehicles (IoV) networks has enabled intelligent transportation and smart traffic management. Nonetheless, the security against adversarial poisoning attacks is also increasingly becoming a challenging task. Specifically, Deep Reinforcement Learning (DRL) is one of the widely used ML designs in IoV applications. The standard ML security techniques are not effective in DRL where the algorithm learns to solve sequential decision-making through continuous interaction with the environment, and the environment is time-varying, dynamic, and mobile. In this paper, we propose a Gated Recurrent Unit (GRU)-based federated continual learning (GFCL) anomaly detection framework against Sybil-based data poisoning attacks in IoV. The objective is to present a lightweight and scalable framework that learns and detects the illegitimate behavior without having a-priori training dataset consisting of attack samples. We use GRU to predict a future data sequence to analyze and detect illegitimate behavior from vehicles in a federated learning-based distributed manner. We investigate the performance of our framework using real-world vehicle mobility traces. The results demonstrate the effectiveness of our proposed solution in terms of different performance metrics.

Manit Baser, Alperen Yildiz, Dinil Mon Divakaran et al.

The static knowledge representations of large language models (LLMs) inevitably become outdated or incorrect over time. While model-editing techniques offer a promising solution by modifying a model's factual associations, they often produce unpredictable ripple effects, which are unintended behavioral changes that propagate even to the hidden space. In this work, we introduce CLaRE, a lightweight representation-level technique to identify where these ripple effects may occur. Unlike prior gradient-based methods, CLaRE quantifies entanglement between facts using forward activations from a single intermediate layer, avoiding costly backward passes. To enable systematic study, we prepare and analyse a corpus of 11,427 facts drawn from three existing datasets. Using CLaRE, we compute large-scale entanglement graphs of this corpus for multiple models, capturing how local edits propagate through representational space. These graphs enable stronger preservation sets for model editing, audit trails, efficient red-teaming, and scalable post-edit evaluation. In comparison to baselines, CLaRE achieves an average of 62.2% improvement in Spearman correlation with ripple effects while being $2.74\times$ faster, and using $2.85\times$ less peak GPU memory. Besides, CLaRE requires only a fraction of the storage needed by the baselines to compute and preserve fact representations. Our entanglement graphs and corpus are available at https://anonymous.4open.science/r/CLaRE-488E.

Poorvi Joshi, Alakesh Kalita, Mohan Gurusamy

Unmanned Aerial Vehicles (UAVs) are integral in various sectors like agriculture, surveillance, and logistics, driven by advancements in 5G. However, existing research lacks a comprehensive approach addressing both data freshness and security concerns. In this paper, we address the intricate challenges of data freshness, and security, especially in the context of eavesdropping and jamming in modern UAV networks. Our framework incorporates exponential AoI metrics and emphasizes secrecy rate to tackle eavesdropping and jamming threats. We introduce a transformer-enhanced Deep Reinforcement Learning (DRL) approach to optimize task offloading processes. Comparative analysis with existing algorithms showcases the superiority of our scheme, indicating its promising advancements in UAV network management.

Poorvi Joshi, Mohan Gurusamy

This paper introduces a novel approach to time series classification using a Markov Transition Field (MTF)-aided Transformer model, specifically designed for Software-Defined Networks (SDNs). The proposed model integrates the temporal dependency modeling strengths of MTFs with the sophisticated pattern recognition capabilities of Transformer architectures. We evaluate the model's performance using the InSDN dataset, demonstrating that our model outperforms baseline classification models, particularly in data-constrained environments commonly encountered in SDN applications. We also highlight the relationship between the MTF and Transformer components, which leads to better performance, even with limited data. Furthermore, our approach achieves competitive training and inference times, making it an efficient solution for real-world SDN applications. These findings establish the potential of MTF-aided Transformers to address the challenges of time series classification in SDNs, offering a promising path for reliable and scalable analysis in scenarios with sparse data.

Manit Baser, Dinil Mon Divakaran, Mohan Gurusamy

Robust model-editing techniques are essential for deploying large language models (LLMs) in practical applications, to enable cost-effective ways to deal with challenges such as privacy breaches, bias mitigation and misinformation spread. For example, an LLM-based healthcare assistance may need to update out-dated or incorrect knowledge to prevent harmful recommendations. However, many editing techniques focus on isolated facts, which critically fail to prevent indirect knowledge leakage -- the unintended reconstruction of edited-out information through persistent causal links and contextual relationships. To assist users in selecting the right editing technique, we develop and present ThinkEval, a framework to systematically quantify indirect knowledge leakage and ripple effects in model-editing. ThinkEval builds and employs specialized knowledge graphs to analyze the causal structure of facts before and after editing. To support this approach, we present KnowGIC, a benchmark dataset comprising multi-step reasoning paths that precisely measure these complex knowledge transformation effects. We evaluate five editing techniques: AlphaEdit, RECT, ROME, MEMIT, and PRUNE across multiple LLMs. Our results show that these techniques struggle to balance indirect fact suppression with the preservation of related knowledge, compromising the contextual integrity of a model's knowledge. Our dataset is available at: https://anonymous.4open.science/r/KnowGIC.

Binghui Wu, Dinil Mon Divakaran, Mohan Gurusamy

As modern networks grow increasingly complex--driven by diverse devices, encrypted protocols, and evolving threats--network traffic analysis has become critically important. Existing machine learning models often rely only on a single representation of packets or flows, limiting their ability to capture the contextual relationships essential for robust analysis. Furthermore, task-specific architectures for supervised, semi-supervised, and unsupervised learning lead to inefficiencies in adapting to varying data formats and security tasks. To address these gaps, we propose UniNet, a unified framework that introduces a novel multi-granular traffic representation (T-Matrix), integrating session, flow, and packet-level features to provide comprehensive contextual information. Combined with T-Attent, a lightweight attention-based model, UniNet efficiently learns latent embeddings for diverse security tasks. Extensive evaluations across four key network security and privacy problems--anomaly detection, attack classification, IoT device identification, and encrypted website fingerprinting--demonstrate UniNet's significant performance gain over state-of-the-art methods, achieving higher accuracy, lower false positive rates, and improved scalability. By addressing the limitations of single-level models and unifying traffic analysis paradigms, UniNet sets a new benchmark for modern network security.

Anum Talpur, Mohan Gurusamy

Machine learning (ML) has made incredible impacts and transformations in a wide range of vehicular applications. As the use of ML in Internet of Vehicles (IoV) continues to advance, adversarial threats and their impact have become an important subject of research worth exploring. In this paper, we focus on Sybil-based adversarial threats against a deep reinforcement learning (DRL)-assisted IoV framework and more specifically, DRL-based dynamic service placement in IoV. We carry out an experimental study with real vehicle trajectories to analyze the impact on service delay and resource congestion under different attack scenarios for the DRL-based dynamic service placement application. We further investigate the impact of the proportion of Sybil-attacked vehicles in the network. The results demonstrate that the performance is significantly affected by Sybil-based data poisoning attacks when compared to adversary-free healthy network scenario.

Anum Talpur, Mohan Gurusamy

The growth of 5G and edge computing has enabled the emergence of Internet of Vehicles. It supports different types of services with different resource and service requirements. However, limited resources at the edge, high mobility of vehicles, increasing demand, and dynamicity in service request-types have made service placement a challenging task. A typical static placement solution is not effective as it does not consider the traffic mobility and service dynamics. Handling dynamics in IoV for service placement is an important and challenging problem which is the primary focus of our work in this paper. We propose a Deep Reinforcement Learning-based Dynamic Service Placement (DRLD-SP) framework with the objective of minimizing the maximum edge resource usage and service delay while considering the vehicle's mobility, varying demand, and dynamics in the requests for different types of services. We use SUMO and MATLAB to carry out simulation experiments. The experimental results show that the proposed DRLD-SP approach is effective and outperforms other static and dynamic placement approaches.

Anum Talpur, Mohan Gurusamy

Machine Learning (ML) has emerged as an attractive and viable technique to provide effective solutions for a wide range of application domains. An important application domain is vehicular networks wherein ML-based approaches are found to be very useful to address various problems. The use of wireless communication between vehicular nodes and/or infrastructure makes it vulnerable to different types of attacks. In this regard, ML and its variants are gaining popularity to detect attacks and deal with different kinds of security issues in vehicular communication. In this paper, we present a comprehensive survey of ML-based techniques for different security issues in vehicular networks. We first briefly introduce the basics of vehicular networks and different types of communications. Apart from the traditional vehicular networks, we also consider modern vehicular network architectures. We propose a taxonomy of security attacks in vehicular networks and discuss various security challenges and requirements. We classify the ML techniques developed in the literature according to their use in vehicular network applications. We explain the solution approaches and working principles of these ML techniques in addressing various security challenges and provide insightful discussion. The limitations and challenges in using ML-based methods in vehicular networks are discussed. Finally, we present observations and lessons learned before we conclude our work.

Anum Talpur, Mohan Gurusamy

The emergence of technologies such as 5G and mobile edge computing has enabled provisioning of different types of services with different resource and service requirements to the vehicles in a vehicular network.The growing complexity of traffic mobility patterns and dynamics in the requests for different types of services has made service placement a challenging task. A typical static placement solution is not effective as it does not consider the traffic mobility and service dynamics. In this paper, we propose a reinforcement learning-based dynamic (RL-Dynamic) service placement framework to find the optimal placement of services at the edge servers while considering the vehicle's mobility and dynamics in the requests for different types of services. We use SUMO and MATLAB to carry out simulation experiments. In our learning framework, for the decision module, we consider two alternative objective functions-minimizing delay and minimizing edge server utilization. We developed an ILP based problem formulation for the two objective functions. The experimental results show that 1) compared to static service placement, RL-based dynamic service placement achieves fair utilization of edge server resources and low service delay, and 2) compared to delay-optimized placement, server utilization optimized placement utilizes resources more effectively, achieving higher fairness with lower edge-server utilization.

Biswadeep Chakraborty, Dinil Mon Divakaran, Ido Nevat et al.

Classification of IoT devices into different types is of paramount importance, from multiple perspectives, including security and privacy aspects. Recent works have explored machine learning techniques for fingerprinting (or classifying) IoT devices, with promising results. However, existing works have assumed that the features used for building the machine learning models are readily available or can be easily extracted from the network traffic; in other words, they do not consider the costs associated with feature extraction. In this work, we take a more realistic approach, and argue that feature extraction has a cost, and the costs are different for different features. We also take a step forward from the current practice of considering the misclassification loss as a binary value, and make a case for different losses based on the misclassification performance. Thereby, and more importantly, we introduce the notion of risk for IoT device classification. We define and formulate the problem of cost-aware IoT device classification. This being a combinatorial optimization problem, we develop a novel algorithm to solve it in a fast and effective way using the Cross-Entropy (CE) based stochastic optimization technique. Using traffic of real devices, we demonstrate the capability of the CE based algorithm in selecting features with minimal risk of misclassification while keeping the cost for feature extraction within a specified limit.

Srinikethan Madapuzi Srinivasan, Tram Truong-Huu, Mohan Gurusamy

With the proliferation of network devices and rapid development in information technology, networks such as Internet of Things are increasing in size and becoming more complex with heterogeneous wired and wireless links. In such networks, link faults may result in a link disconnection without immediate replacement or a link reconnection, e.g., a wireless node changes its access point. Identifying whether a link disconnection or a link reconnection has occurred and localizing the failed link become a challenging problem. An active probing approach requires a long time to probe the network by sending signaling messages on different paths, thus incurring significant communication delay and overhead. In this paper, we adopt a passive approach and develop a three-stage machine learning-based technique, namely ML-LFIL that identifies and localizes link faults by analyzing the measurements captured from the normal traffic flows, including aggregate flow rate, end-to-end delay and packet loss. ML-LFIL learns the traffic behavior in normal working conditions and different link fault scenarios. We train the learning model using support vector machine, multi-layer perceptron and random forest. We implement ML-LFIL and carry out extensive experiments using Mininet platform. Performance studies show that ML-LFIL achieves high accuracy while requiring much lower fault localization time compared to the active probing approach.

Akash Raj Narayanadoss, Tram Truong-Huu, Purnima Murali Mohan et al.

Recent developments in intelligent transport systems (ITS) based on smart mobility significantly improves safety and security over roads and highways. ITS networks are comprised of the Internet-connected vehicles (mobile nodes), roadside units (RSU), cellular base stations and conventional core network routers to create a complete data transmission platform that provides real-time traffic information and enable prediction of future traffic conditions. However, the heterogeneity and complexity of the underlying ITS networks raise new challenges in intrusion prevention of mobile network nodes and detection of security attacks due to such highly vulnerable mobile nodes. In this paper, we consider a new type of security attack referred to as crossfire attack, which involves a large number of compromised nodes that generate low-intensity traffic in a temporally coordinated fashion such that target links or hosts (victims) are disconnected from the rest of the network. Detection of such attacks is challenging since the attacking traffic flows are indistinguishable from the legitimate flows. With the support of software-defined networking that enables dynamic network monitoring and traffic characteristic extraction, we develop a machine learning model that can learn the temporal correlation among traffic flows traversing in the ITS network, thus differentiating legitimate flows from coordinated attacking flows. We use different deep learning algorithms to train the model and study the performance using Mininet-WiFi emulation platform. The results show that our approach achieves a detection accuracy of at least 80%.

Rishi Sairam, Suman Sankar Bhunia, Vijayanand Thangavelu et al.

This is the era of smart devices or things which are fueling the growth of Internet of Things (IoT). It is impacting every sphere around us, making our life dependent on this technological feat. It is of high concern that these smart things are being targeted by cyber criminals taking advantage of heterogeneity, minuscule security features and vulnerabilities within these devices. Conventional centralized IT security measures have limitations in terms of scalability and cost. Therefore, these smart devices are required to be monitored closer to their location ideally at the edge of IoT networks. In this paper, we explore how some security features can be implemented at the network edge to secure these smart devices. We explain the importance of Network Function Virtualization (NFV) in order to deploy security functions at the network edge. To achieve this goal, we introduce NETRA - a novel lightweight Docker-based architecture for virtualizing network functions to provide IoT security. Also, we highlight the advantages of the proposed architecture over the standardized NFV architecture in terms of storage, memory usage, latency, throughput, load average, scalability and explain why the standardized architecture is not suitable for IoT. We study the performance of proposed NFV based edge analysis for IoT security and show that attacks can be detected with more than 95% accuracy in less than a second.