Enterprise intrusion response still depends on static playbooks and analyst-driven triage, creating delay between alert generation and containment. We present Agentra, a supervisable multi-agent Intrusion Response System (IRS) framework that converts alerts from IDS, EDR, and XDR platforms into structured incident response plans grounded in MITRE ATT CK, MITRE D3FEND, and NIST CSF 2.0. Agentra decomposes response reasoning across role-scoped agents, validates proposed plans through a bounded Planner–Validator review loop, screens retrieved threat intelligence through a Moderator security gateway, gates actions through an Action Catalog and risk score, and records decisions in an append-only audit log. We evaluate Agentra against a static OASIS CACAO v2.0 cyber-playbook baseline on a 120-event corpus drawn from ThreatHunter-Playbook, Splunk BOTSv3, and DARPA OpTC. The strongest configuration improves FP-aware IRS F1 from 0.61 to 0.84 and restores the projected harmful-action rate to the static baseline level of 0.0

Large Language Models (LLMs) are increasingly fine-tuned for critical-domain Question-Answering (QA), yet choosing which small model to adapt, before paying the cost of adaptation, remains difficult. Fine-tuning can improve domain alignment, but it may also erode prior knowledge, weaken instruction-following, or increase hallucination, especially when labeled data are scarce or rapidly evolving as in cybersecurity. We present FiT (Find before Fine-Tune), a task-oriented diagnostic framework that characterizes small LLMs along three capabilities required for cybersecurity QA: vocabulary recognition, parametric knowledge, and contextualization of retrieved information. Using FiT, we conduct an empirical study of five open-weight 7-billion-parameter models under two fine-tuning regimes. We find that fine-tuning does not uniformly help: it consistently degrades vocabulary and parametric knowledge in small models, and the two regimes trade off differently. Knowledge-focused tuning causes moderate, rank-preserving degradation, whereas instruction-focused tuning collapses measured knowledge through induced abstention, inverting the knowledge ranking while leaving retrieval-grounded contextualization essentially intact. We quantify these regime-specific patterns with rank-correlation analysis and show that pre-fine-tuning FiT scores anticipate the direction of post-tuning change. Our results suggest that task-oriented diagnosis can screen out unsuitable models, avoid unnecessary fine-tuning, and support safer deployment of small LLMs in cybersecurity QA pipelines.

The digital age has completely transformed the way that information is processed and stored, which makes cybersecurity a crucial field of research. Cybersecurity contains many different domains, but this work focuses on Intrusion Detection Systems (IDSs). Within the literature, Hierarchical Self-Organizing Maps (HSOMs) have been used to create trustworthy, explainable, and AI-based IDSs. However, HSOMs are trained sequentially, which means that training HSOMs on large datasets is slow. This work presents a novel parallel HSOM architecture, called parHSOM. The purpose of this research is to investigate the effect that parallel computation has on the HSOM training time. parHSOM is tested on two different testbeds, four different output grid sizes, and five different cybersecurity datasets. Performance metrics collected from these experiments show that parHSOM consistently trains faster than the Sequential HSOM algorithm without any significant loss in performance. Additionally, this work provides a platform for further investigation into parallel HSOM implementations.

Watershed networks exhibit convergent topologies in which multiple tributaries merge into downstream channels,integrating diverse upstream hydrological processes. In ungauged basins, the absence of direct observations increases uncertainty and limits the ability to anticipate extreme events. This study evaluates whether an encoder-only Transformer provides an advantage over an LSTM for upstream streamflow inference under limited hydrologic information, using retrospective simulations from the NOAA National Water Model (NWM). Across both upstream-only and combined configurations, the LSTM showed stronger overall performance than the Transformer model across the two configurations. Incorporating downstream information further boosted performance for all models, increasing median NNSE by more than 60

Adversaries continuously evolve their tactics, techniques, and procedures (TTPs) to achieve their objectives while evading detection, requiring defenders to continually update their understanding of adversary behavior. Prior research has proposed automated extraction of TTP-related intelligence from unstructured text and mapping it to structured knowledge bases, such as MITRE ATT CK. However, existing work varies widely in extraction objectives, datasets, modeling approaches, and evaluation practices, making it difficult to understand the research landscape. The goal of this study is to aid security researchers in understanding the state of the art in extracting attack tactics, techniques, and procedures (TTPs) from unstructured text by analyzing relevant literature. We systematically analyze 80 peer-reviewed studies across key dimensions: extraction purposes, data sources, dataset construction, modeling approaches, evaluation metrics, and artifact availability. Our analysis reveals several dominant trends. Technique-level classification remains the dominant task formulation, while tactic classification and technique searching are underexplored. The field has progressed from rule-based and traditional machine learning to transformer-based architectures (e.g., BERT, SecureBERT, RoBERTa), with recent studies exploring LLM-based approaches including prompting, retrieval-augmented generation, and fine-tuning, though adoption remains emergent. Despite these advances, important limitations persist: many studies rely on single-label classification, limited evaluation settings, and narrow datasets, constraining cross-domain generalization. Reproducibility is further hindered by proprietary datasets, limited code releases, and restricted corpora.

Multi-agent systems (MAS) powered by LLMs promise adaptive, reasoning-driven enterprise workflows, yet granting agents autonomous control over tools, memory, and communication introduces attack surfaces absent from deterministic pipelines. While current research largely addresses prompt-level exploits and narrow individual vectors, it lacks a holistic architectural model for enterprise-grade security. We introduce AgenticCyOps (Securing Multi-Agentic AI Integration in Enterprise Cyber Operations), a framework built on a systematic decomposition of attack surfaces across component, coordination, and protocol layers, revealing that documented vectors consistently trace back to two integration surfaces: tool orchestration and memory management. Building on this observation, we formalize these integration surfaces as primary trust boundaries and define five defensive principles: authorized interfaces, capability scoping, verified execution, memory integrity synchronization, and access-controlled data isolation; each aligned with established compliance standards (NIST, ISO 27001, GDPR, EU AI Act). We apply the framework to a Security Operations Center (SOC) workflow, adopting the Model Context Protocol (MCP) as the structural basis, with phase-scoped agents, consensus validation loops, and per-organization memory boundaries. Coverage analysis, attack path tracing, and trust boundary assessment confirm that the design addresses the documented attack vectors with defense-in-depth, intercepts three of four representative attack chains within the first two steps, and reduces exploitable trust boundaries by a minimum of 72

Sycophancy and hallucination are persistent failure modes of Large Language Models (LLMs) across domains. However, it becomes particularly consequential in clinical question answering, where responses must remain grounded in the provided context and robust to user pressure. Hallucination can introduce information that is unsupported by the context, while sycophancy can cause a model to abandon a previously correct answer when challenged by the user. Existing approaches, such as prompt-based safeguards and always-on activation steering, often address these behaviors separately or apply interventions broadly across turns, which can unnecessarily deteriorate responses that were already correct. To address these limitations within a single framework, we employ Inference Time Intervention (ITI) to jointly control both behaviors by learning separate steering directions for hallucination and sycophancy from contrastive clinical pairs and applying them to causally verified attention heads. During runtime, behavior-specific gates then determine when intervention is needed: the hallucination component mitigates unsupported claims, while the sycophancy component mitigates answer shifts caused by user pressure. We evaluate this framework on clinical questions grounded in EHR data while keeping the model weights frozen. Across all evaluation settings, we conducted 15,900 model-response runs. Across 600 pressure trajectories for the 4-billion-parameter model, the unsteered model caved in 570 cases. At the same time, gated steering helped it last longer in 551 of them. It held its ground under pressure at levels comparable to those of models with more than 100 billion parameters, showing that targeted inference-time steering can improve robustness without intervening at every turn.

Cognitive AI seeks to move beyond language generation and autonomous task execution toward systems capable of sustained reasoning, adaptive behavior, persistent memory, and self-regulation. While generative and agentic AI have demonstrated impressive capabilities across a wide range of tasks, many fundamental cognitive functions remain fragmented or weakly developed, limiting reliable operation over extended time horizons. This paper presents a taxonomy-driven survey of the major cognitive capability gaps that continue to constrain the development of Cognitive AI. The literature is organized around five dimensions: persistent state modeling, goal-directed autonomy, self-monitoring and control, environment interaction, and learning and adaptation. For each dimension, we review recent advances, identify recurring limitations, and discuss open research challenges. Building on these insights, we outline a conceptual Adaptive Cognitive Intelligence Architecture (ACIA) and examine emerging directions in cognition-centric evaluation. The proposed taxonomy provides a unified framework for organizing existing research, identifying unresolved challenges, and guiding the design of future cognitively capable systems. Together, the taxonomy, architectural perspective, and evaluation framework offer a roadmap for advancing AI systems that exhibit more reliable long-term reasoning, adaptive decision-making, and continual learning. The survey highlights key research opportunities toward more adaptive, reliable, and cognitively capable AI systems, providing a foundation for future progress toward Cognitive AI and, ultimately, Artificial General Intelligence (AGI).

Modern enterprise systems face escalating cyber threats that are increasingly dynamic, distributed, and multi-stage in nature. Traditional intrusion detection and response systems often rely on static rules and manual workflows, which limit their ability to respond with the speed and precision required in high-stakes environments. To address these challenges, we present the Intrusion Response System Digital Assistant (IRSDA), an agent-based framework designed to deliver autonomous and policy-compliant cyber defense. IRSDA combines Self-Adaptive Autonomic Computing Systems (SA-ACS) with the Knowledge guided Monitor, Analyze, Plan, and Execute (MAPE-K) loop to support real-time, partition-aware decision-making across enterprise infrastructure. IRSDA incorporates a knowledge-driven architecture that integrates contextual information with AI-based reasoning to support system-guided intrusion response. The framework leverages retrieval mechanisms and structured representations to inform decision-making while maintaining alignment with operational policies. We assess the system using a representative real-world microservices application, demonstrating its ability to automate containment, enforce compliance, and provide traceable outputs for security analyst interpretation. This work outlines a modular and agent-driven approach to cyber defense that emphasizes explainability, system-state awareness, and operational control in intrusion response.

Vehicles are complex Cyber Physical Systems (CPS) that operate in a variety of environments, and the likelihood of failure of one or more subsystems, such as the engine, transmission, brakes, and fuel, can result in unscheduled downtime and incur high maintenance or repair costs. In order to prevent these issues, it is crucial to continuously monitor the health of various subsystems and identify abnormal sensor channel behavior. Data-driven Digital Twin (DT) systems are capable of such a task. Current DT technologies utilize various Deep Learning (DL) techniques that are constrained by the lack of justification or explanation for their predictions. This inability of these opaque systems can influence decision-making and raises user trust concerns. This paper presents a solution to this issue, where the TwinExplainer system, with its three-layered architectural pipeline, explains the predictions of an automotive DT. Such a system can assist automotive stakeholders in understanding the global scale of the sensor channels and how they contribute towards generic DT predictions. TwinExplainer can also visualize explanations for both normal and abnormal local predictions computed by the DT.

This paper presents CLINICSUM, a novel framework designed to automatically generate clinical summaries from patient-doctor conversations. It utilizes a two-module architecture: a retrieval-based filtering module that extracts Subjective, Objective, Assessment, and Plan (SOAP) information from conversation transcripts, and an inference module powered by fine-tuned Pre-trained Language Models (PLMs), which leverage the extracted SOAP data to generate abstracted clinical summaries. To fine-tune the PLM, we created a training dataset of consisting 1,473 conversations-summaries pair by consolidating two publicly available datasets, FigShare and MTS-Dialog, with ground truth summaries validated by Subject Matter Experts (SMEs). CLINICSUM’s effectiveness is evaluated through both automatic metrics (e.g., ROUGE, BERTScore) and expert human assessments. Results show that CLINICSUM outperforms state-of-the-art PLMs, demonstrating superior precision, recall, and F-1 scores in automatic evaluations and receiving high preference from SMEs in human assessment, making it a robust solution for automated clinical summarization.

The adoption of Machine Learning Operations (MLOps) enables automation and reliable model deployments across industries. However, differing MLOps lifecycle frameworks and maturity models proposed by industry, academia, and organizations have led to confusion regarding standard adoption practices. This paper introduces a unified MLOps lifecycle framework, further incorporating Large Language Model Operations (LLMOps), to address this gap. Additionally, we outlines key roles, tools, and costs associated with MLOps adoption at various maturity levels. By providing a standardized framework, we aim to help organizations clearly define and allocate the resources needed to implement MLOps effectively.

This paper addresses the critical challenge of estimating the reliability of an Electric Vehicle (EV) charging systems when facing risks such as overheating, unpredictable, weather, and cyberattacks. Traditional methods for predicting failures often rely on past data or limiting assumptions, making them ineffective for new or less common threats that results in failure. To solve this issue, we utilize the Principle of Maximum Entropy (PME), a statistical tool that estimates risks even with limited information. PME works by balancing known constraints to create an unbiased predictions without guessing missing details. Using the EV charging ecosystem as a case study, we show how PME models stress factors responsible for failure. Our findings reveal a critical insight: even minor, localized stress events can trigger disproportionately large drops in overall system reliability, similar to a domino effect. The our PME model demonstrates how high-impact components, such as the power grid, are more likely to fail as stress accumulates, creating network-wide tipping points. Beyond EVs, this approach applies to any complex system with incomplete data, such as smart grids, healthcare devices, or logistics networks. By mathematically establishing an inverse relationship between uncertainty (entropy) and reliability, our work quantifies how greater system unpredictability directly degrades robustness. This offers a universal tool to improve decision-making under unpredictable conditions. This work bridges advanced mathematics with real-world engineering, providing actionable insights for policymakers and industries to build safer, more efficient systems in our increasingly connected world.

Intrusion Detection Systems (IDS) are essential for protecting network infrastructure from malicious activities. However, the effectiveness of AI-driven IDS is often compromised by the scarcity of labeled data and high false-alarm rates. Semi-Supervised Learning (SSL) offers a potential solution to the data labeling problem, but existing SSL approaches have shown limited success in the IDS domain. In this paper, we propose a novel Artificial Immune System-based IDS (AIS-IDS) architecture. We instantiate this architecture as SMARTI, an SSL-based AIS-IDS designed to operate effectively without labeled threat data and to minimize false positives by leveraging the self/non-self model of immunology. Our experimental results demonstrate that SMARTI achieves up to 96.7

Security Operations Center (SoC) analysts gather threat reports from openly accessible global threat repositories and tailor the information to their organization's needs, such as developing threat intelligence and security policies. They also depend on organizational internal repositories, which act as private local knowledge database. These local knowledge databases store credible cyber intelligence, critical operational and infrastructure details. SoCs undertake a manual labor-intensive task of utilizing these global threat repositories and local knowledge databases to create both organization-specific threat intelligence and mitigation policies. Recently, Large Language Models (LLMs) have shown the capability to process diverse knowledge sources efficiently. We leverage this ability to automate this organization-specific threat intelligence generation. We present LocalIntel, a novel automated threat intelligence contextualization framework that retrieves zero-day vulnerability reports from the global threat repositories and uses its local knowledge database to determine implications and mitigation strategies to alert and assist the SoC analyst. LocalIntel comprises two key phases: knowledge retrieval and contextualization. Quantitative and qualitative assessment has shown effectiveness in generating up to 93% accurate organizational threat intelligence with 64% inter-rater agreement.

Signature-based Intrusion Detection Systems (IDS) detect malicious activity by matching network or host events against predefined rules. Security analysts manually develop these rules from Cyber Threat Intelligence (CTI). As threats evolve, this manual pipeline faces two bottlenecks. Before authoring a new rule, an analyst must reconcile the incoming CTI with the existing rule base and determine whether to create, update, or retire one. This process is challenging due to the representational differences between the CTI and Rule formats. This gap limits the effectiveness of keyword- and embedding-based search, making rule reconciliation cognitively demanding and, in turn, contributing to "rule bloat". Second, automated verification of a new rule is inherently difficult as zero-day threats lack ground truth from simulated testing. Hence, standard metrics cannot prove that a rule semantically adheres to the CTI, and the use of LLMs leads to non-deterministic behavior. To address these challenges, we introduce FALCON, an agentic framework for CTI-grounded rule retrieval, generation, and validation. At its core, a novel CTI-Rule semantic scorer, quantifies the functional alignment between a CTI and a rule; the same signal drives a retriever that surfaces relevant deployed rules and a ground-truth-free validator that scores generated ones. Around it, a generation pipeline produces deployable rules from CTI in real time and refines them through self-reflective syntactic, semantic, and performance validators. Across network (Snort) and host-based (YARA) platforms on a purpose-built CTI-Rule dataset, FALCON attains a mean relevance of 0.72 (approx), with 84

Providing contextual and comprehensive medical information tailored to individual patients is critical for enabling effective care in the healthcare domain. However, existing approaches often struggle to deliver personalized responses due to the distributed nature of medical data across multiple sources such as patient records, medical literature, and online resources. To address this challenge, we present MedInsight, a multi-source context augmentation framework that leverages Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) techniques to augment patient-specific information from medical transcripts with trusted knowledge from textbooks and web resources to generate personalized and contextually relevant responses. Our framework consists of three phases: patient context retrieval, medical knowledge retrieval, and response generation. By augmenting patient context with relevant external knowledge, MedInsight generates contextually relevant responses, empowering patients and caregivers with actionable insights. Experiments on the MTSamples dataset validate MedInsight’s effectiveness in generating contextually appropriate medical responses, using a comprehensive set of metrics including RAGAs, TruLens, ROUGE, and BertScore. Additionally, qualitative evaluations by Subject-Matter Experts (SMEs) further confirm the relevance and factual correctness of the generated responses.

The shift toward patient-centric healthcare requires understanding comprehensive patient journeys. Current healthcare data systems often fail to provide holistic representations, hindering coordinated care. Patient Journey Knowledge Graphs (PJKGs) solve this by integrating diverse patient information into unified, structured formats. This paper presents a methodology for constructing PJKGs using Large Language Models (LLMs) to process both clinical documentation and patient-provider conversations. These graphs capture temporal and causal relationships between clinical events, enabling advanced reasoning and personalized insights. Our evaluation of four LLMs (Claude 3.5, Mistral, Llama 3.1, ChatGPT4o) shows all achieved perfect structural compliance but varied in medical entity processing, computational efficiency, and semantic accuracy. This work advances patient-centric healthcare through actionable knowledge graphs (KGs) that enhance care coordination and outcome prediction.

Anomaly detection in complex industrial environments poses unique challenges, particularly in contexts characterized by data sparsity and evolving operational conditions. Predictive maintenance (PdM) in such settings demands methodologies that are adaptive, transferable, and capable of integrating domain-specific knowledge. In this paper, we present RAAD-LLM, a novel framework for adaptive anomaly detection, leveraging large language models (LLMs) integrated with Retrieval-Augmented Generation (RAG). This approach addresses the aforementioned PdM challenges. By effectively utilizing domain-specific knowledge, RAAD-LLM enhances the detection of anomalies in time series data without requiring fine-tuning on specific datasets. The framework's adaptability mechanism enables it to adjust its understanding of normal operating conditions dynamically, thus increasing detection accuracy. We validate this methodology through a real-world application for a plastics manufacturing plant and the Skoltech Anomaly Benchmark (SKAB). Results show significant improvements over our previous model with an accuracy increase from 70.7 to 89.1 on the real-world dataset. By allowing for the enriching of input series data with semantics, RAAD-LLM incorporates multimodal capabilities that facilitate more collaborative decision-making between the model and plant operators. Overall, our findings support RAAD-LLM's ability to revolutionize anomaly detection methodologies in PdM, potentially leading to a paradigm shift in how anomaly detection is implemented across various industries.