arXiv subjects

Md Jueal Mia

Publications and source records attributed to Md Jueal Mia.

10 recordsLinked to original sources

arXiv2026

SoK: Rethinking Jailbreaking in the Era of Agentic AI: Attacks, Defenses, and Practical Consideration

Large language models (LLMs) are rapidly evolving from conversational assistants into agentic AI systems that reason, plan, invoke tools, maintain persistent memory, communicate with other agents, and execute multi-step tasks. At the same time, modern models exhibit substantially stronger native safety alignment than earlier generations on which many jailbreak attacks and defenses were originally studied. This shift raises a fundamental question: \textit{which established jailbreak-security findings remain valid in the era of modern LLMs and agentic AI?} We address this question through a Systematization of Knowledge (SoK) that reframes jailbreak security around the full agentic execution pipeline. We develop unified taxonomies of attacks and defenses spanning user interaction, planning and reasoning, memory, tool use, and inter-agent communication, and introduce a security--utility--efficiency evaluation framework that separates native harmful-prompt safety, adversarial jailbreak robustness, and agent-level security outcomes. We further conduct a controlled empirical study of representative attacks and defenses within a common agentic framework. Our results reveal three important gaps. First, strong native alignment does not imply robustness to adversarial jailbreaks. Second, defense effectiveness is highly model-, attack-, and component-dependent and can come at substantial cost in over-refusal, utility, and latency. Third, low final-response attack success can mask severe intermediate compromise: planning, memory, and tool interactions may remain unsafe even when the final response is successfully filtered. These findings motivate a shift from response-centric jailbreak defense toward cross-layer, execution-aware security that protects agent state, component transitions, and external actions while preserving practical utility and efficiency.

cs.AI↗

arXiv2026

Don't Overthink, Don't Underthink: Toward Adaptive Reasoning in Agentic AI

Recent advances in Large Language Models (LLMs) have shown that increased inference-time reasoning can improve performance on complex tasks. However, many existing approaches rely on fixed or preallocated reasoning controls, such as fixed token budgets, pre-execution difficulty estimates, or activation-space interventions, and are often evaluated on standalone reasoning benchmarks rather than full agentic workflows. These assumptions may not hold in agentic AI systems, where reasoning requirements evolve dynamically through planning, tool use, memory retrieval, and agent-to-agent interactions. Consequently, reasoning can become either excessive or insufficient, resulting in unnecessary computation, increased latency, planning drift, excessive tool use, or incomplete solutions. We argue that a major challenge for next-generation agentic AI is not merely how much reasoning a language model should perform, but how it should allocate reasoning according to evolving task demands. We characterize over-reasoning and under-reasoning as recurring failure modes of misallocated reasoning and evaluate them on MATH-500 and the GAIA public validation benchmark. Using tool-decision latency, token consumption, token-limit exhaustion, and answer correctness, our results suggest that cases classified as over-reasoning are associated with higher computational cost without proportional accuracy gains, whereas cases classified as under-reasoning are consistently associated with incorrect or incomplete solutions. These findings motivate future research on adaptive reasoning mechanisms for agentic AI.

cs.AI↗

arXiv2026

FedShield-LLM: A Secure and Scalable Federated Fine-Tuned Large Language Model

Federated Learning (FL) offers a decentralized framework for training and fine-tuning Large Language Models (LLMs) by leveraging computational resources across organizations while keeping sensitive data on local devices. It addresses privacy and security concerns while navigating challenges associated with the substantial computational demands of LLMs, which can be prohibitive for small and medium-sized organizations. FL supports the development of task-specific LLMs for cross-silo applications through fine-tuning but remains vulnerable to inference-related risks that threaten sensitive information. Prior studies have utilized Differential Privacy (DP) in LLM fine-tuning, which, despite being effective at preserving privacy, can degrade model performance. To overcome these challenges, we propose FedShield-LLM which integrates pruning with Fully Homomorphic Encryption (FHE) applied to Low-Rank Adaptation (LoRA) parameters. This combination enables secure computation over encrypted model updates and reduces the attack surface by deactivating less important LoRA parameters. Furthermore, optimized federated algorithms for cross-silo environments enhance scalability and efficiency. Parameter-efficient fine-tuning techniques like LoRA substantially reduce computational and communication overhead, making FL feasible for resource-constrained clients. Extensive experiments using Llama-2 models (7B and 13B) on four diverse datasets demonstrate that FedShield-LLM achieves superior collaborative performance and system efficiency compared to existing methods, supporting practical deployment across multiple domains.

cs.CR↗

arXiv2026

GUARD-SLM: Token Activation-Based Defense Against Jailbreak Attacks for Small Language Models

Small Language Models (SLMs) are emerging as efficient and economically viable alternatives to Large Language Models (LLMs), offering competitive performance with significantly lower computational costs and latency. These advantages make SLMs suitable for resource-constrained and efficient deployment on edge devices. However, existing jailbreak defenses show limited robustness against heterogeneous attacks, largely due to an incomplete understanding of the internal representations across different layers of language models that facilitate jailbreak behaviors. In this paper, we conduct a comprehensive empirical study on 9 jailbreak attacks across 7 SLMs and 3 LLMs. Our analysis shows that SLMs remain highly vulnerable to malicious prompts that bypass safety alignment. We analyze hidden-layer activations across different layers and model architectures, revealing that different input types form distinguishable patterns in the internal representation space. Based on this observation, we propose GUARD-SLM, a lightweight token activation-based method that operates in the representation space to filter malicious prompts during inference while preserving benign ones. Our findings highlight robustness limitations across layers of language models and provide a practical direction for secure small language model deployment.

cs.CR↗

arXiv2025

Jailbreaking Large Vision Language Models in Intelligent Transportation Systems

Large Vision Language Models (LVLMs) demonstrate strong capabilities in multimodal reasoning and many real-world applications, such as visual question answering. However, LVLMs are highly vulnerable to jailbreaking attacks. This paper systematically analyzes the vulnerabilities of LVLMs integrated in Intelligent Transportation Systems (ITS) under carefully crafted jailbreaking attacks. First, we carefully construct a dataset with harmful queries relevant to transportation, following OpenAI's prohibited categories to which the LVLMs should not respond. Second, we introduce a novel jailbreaking attack that exploits the vulnerabilities of LVLMs through image typography manipulation and multi-turn prompting. Third, we propose a multi-layered response filtering defense technique to prevent the model from generating inappropriate responses. We perform extensive experiments with the proposed attack and defense on the state-of-the-art LVLMs (both open-source and closed-source). To evaluate the attack method and defense technique, we use GPT-4's judgment to determine the toxicity score of the generated responses, as well as manual verification. Further, we compare our proposed jailbreaking method with existing jailbreaking techniques and highlight severe security risks involved with jailbreaking attacks with image typography manipulation and multi-turn prompting in the LVLMs integrated in ITS.

cs.AI↗

arXiv2025

JaiLIP: Jailbreaking Vision-Language Models via Loss Guided Image Perturbation

Vision-Language Models (VLMs) have remarkable abilities in generating multimodal reasoning tasks. However, potential misuse or safety alignment concerns of VLMs have increased significantly due to different categories of attack vectors. Among various attack vectors, recent studies have demonstrated that image-based perturbations are particularly effective in generating harmful outputs. In the literature, many existing techniques have been proposed to jailbreak VLMs, leading to unstable performance and visible perturbations. In this study, we propose Jailbreaking with Loss-guided Image Perturbation (JaiLIP), a jailbreaking attack in the image space that minimizes a joint objective combining the mean squared error (MSE) loss between clean and adversarial image with the models harmful-output loss. We evaluate our proposed method on VLMs using standard toxicity metrics from Perspective API and Detoxify. Experimental results demonstrate that our method generates highly effective and imperceptible adversarial images, outperforming existing methods in producing toxicity. Moreover, we have evaluated our method in the transportation domain to demonstrate the attacks practicality beyond toxic text generation in specific domain. Our findings emphasize the practical challenges of image-based jailbreak attacks and the need for efficient defense mechanisms for VLMs.

cs.CV↗

arXiv2025

An Empirical Analysis of Secure Federated Learning for Autonomous Vehicle Applications

Federated Learning lends itself as a promising paradigm in enabling distributed learning for autonomous vehicles applications and ensuring data privacy while enhancing and refining predictive model performance through collaborative training on edge client vehicles. However, it remains vulnerable to various categories of cyber-attacks, necessitating more robust security measures to effectively mitigate potential threats. Poisoning attacks and inference attacks are commonly initiated within the federated learning environment to compromise secure system performance. Secure aggregation can limit the disclosure of sensitive information from outsider and insider attackers of the federated learning environment. In this study, our aim is to conduct an empirical analysis on the transportation image dataset (e.g., LISA traffic light) using various secure aggregation techniques and multiparty computation in the presence of diverse categories of cyber-attacks. Multiparty computation serves as a state-of-the-art security mechanism, offering standard privacy for secure aggregation of edge autonomous vehicles local model updates through various security protocols. The presence of adversaries can mislead the autonomous vehicle learning model, leading to the misclassification of traffic lights, and resulting in detrimental impacts. This empirical study explores the resilience of various secure federated learning aggregation techniques and multiparty computation in safeguarding autonomous vehicle applications against various cyber threats during both training and inference times.

cs.DC↗

arXiv2025

Distributed LLMs and Multimodal Large Language Models: A Survey on Advances, Challenges, and Future Directions

Language models (LMs) are machine learning models designed to predict linguistic patterns by estimating the probability of word sequences based on large-scale datasets, such as text. LMs have a wide range of applications in natural language processing (NLP) tasks, including autocomplete and machine translation. Although larger datasets typically enhance LM performance, scalability remains a challenge due to constraints in computational power and resources. Distributed computing strategies offer essential solutions for improving scalability and managing the growing computational demand. Further, the use of sensitive datasets in training and deployment raises significant privacy concerns. Recent research has focused on developing decentralized techniques to enable distributed training and inference while utilizing diverse computational resources and enabling edge AI. This paper presents a survey on distributed solutions for various LMs, including large language models (LLMs), vision language models (VLMs), multimodal LLMs (MLLMs), and small language models (SLMs). While LLMs focus on processing and generating text, MLLMs are designed to handle multiple modalities of data (e.g., text, images, and audio) and to integrate them for broader applications. To this end, this paper reviews key advancements across the MLLM pipeline, including distributed training, inference, fine-tuning, and deployment, while also identifying the contributions, limitations, and future areas of improvement. Further, it categorizes the literature based on six primary focus areas of decentralization. Our analysis describes gaps in current methodologies for enabling distributed solutions for LMs and outline future research directions, emphasizing the need for novel solutions to enhance the robustness and applicability of distributed LMs.

cs.CL↗

arXiv2024

QuanCrypt-FL: Quantized Homomorphic Encryption with Pruning for Secure Federated Learning

Federated Learning has emerged as a leading approach for decentralized machine learning, enabling multiple clients to collaboratively train a shared model without exchanging private data. While FL enhances data privacy, it remains vulnerable to inference attacks, such as gradient inversion and membership inference, during both training and inference phases. Homomorphic Encryption provides a promising solution by encrypting model updates to protect against such attacks, but it introduces substantial communication overhead, slowing down training and increasing computational costs. To address these challenges, we propose QuanCrypt-FL, a novel algorithm that combines low-bit quantization and pruning techniques to enhance protection against attacks while significantly reducing computational costs during training. Further, we propose and implement mean-based clipping to mitigate quantization overflow or errors. By integrating these methods, QuanCrypt-FL creates a communication-efficient FL framework that ensures privacy protection with minimal impact on model accuracy, thereby improving both computational efficiency and attack resilience. We validate our approach on MNIST, CIFAR-10, and CIFAR-100 datasets, demonstrating superior performance compared to state-of-the-art methods. QuanCrypt-FL consistently outperforms existing method and matches Vanilla-FL in terms of accuracy across varying client. Further, QuanCrypt-FL achieves up to 9x faster encryption, 16x faster decryption, and 1.5x faster inference compared to BatchCrypt, with training time reduced by up to 3x.

cs.CR↗

arXiv2024

CorBin-FL: A Differentially Private Federated Learning Mechanism using Common Randomness

Federated learning (FL) has emerged as a promising framework for distributed machine learning. It enables collaborative learning among multiple clients, utilizing distributed data and computing resources. However, FL faces challenges in balancing privacy guarantees, communication efficiency, and overall model accuracy. In this work, we introduce CorBin-FL, a privacy mechanism that uses correlated binary stochastic quantization to achieve differential privacy while maintaining overall model accuracy. The approach uses secure multi-party computation techniques to enable clients to perform correlated quantization of their local model updates without compromising individual privacy. We provide theoretical analysis showing that CorBin-FL achieves parameter-level local differential privacy (PLDP), and that it asymptotically optimizes the privacy-utility trade-off between the mean square error utility measure and the PLDP privacy measure. We further propose AugCorBin-FL, an extension that, in addition to PLDP, achieves user-level and sample-level central differential privacy guarantees. For both mechanisms, we derive bounds on privacy parameters and mean squared error performance measures. Extensive experiments on MNIST and CIFAR10 datasets demonstrate that our mechanisms outperform existing differentially private FL mechanisms, including Gaussian and Laplacian mechanisms, in terms of model accuracy under equal PLDP privacy budgets.

cs.LG↗