Search Results (387)

Large language models (LLMs) now match or exceed human performance on many open-ended language tasks, yet they continue to produce fluent but incorrect statements, which is a failure mode widely referred to as hallucination. In low-stakes settings this may be tolerable; in regulated or safety-critical domains such as financial services, compliance review, and client decision support, it is not. Motivated by these realities, we develop an integrated mitigation framework that layers complementary controls rather than relying on any single technique. The framework combines structured prompt design, retrieval-augmented generation (RAG) with verifiable evidence sources, and targeted fine-tuning aligned with domain truth constraints. Our interest in this problem is practical. Individual mitigation techniques have matured quickly, yet teams deploying LLMs in production routinely report difficulty stitching them together in a coherent, maintainable pipeline. Decisions about when to ground a response in retrieved data, when to escalate uncertainty, how to capture provenance, and how to evaluate fidelity are often made ad hoc. Drawing on experience from financial technology implementations, where even rare hallucinations can carry material cost, regulatory exposure, or loss of customer trust, we aim to provide clearer guidance in the form of an easy-to-follow tutorial. This paper makes four contributions. First, we introduce a three-layer reference architecture that organizes mitigation activities across input governance, evidence-grounded generation, and post-response verification. Second, we describe a lightweight supervisory agent that manages uncertainty signals and triggers escalation (to humans, alternate models, or constrained workflows) when confidence falls below policy thresholds. Third, we analyze common but under-addressed security surfaces relevant to hallucination mitigation, including prompt injection, retrieval poisoning, and policy evasion attacks. Finally, we outline an implementation playbook for production deployment, including evaluation metrics, operational trade-offs, and lessons learned from early financial-services pilots. Full article

Retrieval-Augmented Generation (RAG) combined with Large Language Models (LLMs) introduces a new paradigm for clinical-trial data analysis that is both real-time and knowledge-traceable. This study targets a multi-site, real-world data environment. It builds a hierarchical RAG pipeline spanning an electronic health record (EHR), National Health Insurance (NHI) billing codes, and image-vector indices. The LLM is optimized through lightweight LoRA/QLoRA fine-tuning and reinforcement-learning-based alignment. The system first retrieves key textual and imaging evidence from heterogeneous data repositories and then fuses these artifacts into the contextual window for clinical report generation. Experimental results show marked improvements over traditional manual statistics and prompt-only models in retrieval accuracy, textual coherence, and response latency while reducing human error and workload. In evaluation, the proposed multimodal RAG-LLM workflow achieved statistically significant gains in three core metrics—recall, factual consistency, and expert ratings—and substantially shortened overall report-generation time, demonstrating clear efficiency advantages versus conventional manual processes. However, LLMs alone often face challenges such as limited real-world grounding, hallucination risks, and restricted context windows. Similarly, RAG systems, while improving factual consistency, depend heavily on retrieval quality and may yield incoherent synthesis if evidence is misaligned. These limitations underline the complementary nature of integrating RAG and LLM architectures in a clinical reporting context. Quantitatively, the proposed system achieved a Composite Quality Index (CQI) of 78.3, outperforming strong baselines such as Med-PaLM 2 (72.6) and PMC-LLaMA (74.3), and reducing the report drafting time by over 75% (p < 0.01). These findings confirm the practical feasibility of the framework to support fully automated clinical reporting. Full article

Post-Traumatic Stress Disorder (PTSD) poses complex clinical challenges due to its emotional volatility, contextual sensitivity, and need for personalized care. Conventional AI systems often fall short in therapeutic contexts due to lack of explainability, ethical safeguards, and narrative understanding. We propose a hybrid neuro-symbolic architecture that combines Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), symbolic controllers, and ensemble classifiers to support clinicians in PTSD follow-up. The proposal integrates real-time anonymization, session memory through patient-specific RAG, and a Human-in-the-Loop (HITL) interface. It ensures clinical safety via symbolic logic rules derived from trauma-informed protocols. The proposed architecture enables safe, personalized AI-driven responses by combining statistical language modeling with explicit therapeutic constraints. Through modular integration, it supports affective signal adaptation, longitudinal memory, and ethical traceability. A comparative evaluation against state-of-the-art approaches highlights improvements in contextual alignment, privacy protection, and clinician supervision. Full article

The rapid retrieval and precise acquisition of land use approval information are crucial for enhancing the efficiency and quality of land use approval, as well as for promoting the intelligent transformation of land use approval processes. As an advanced retrieval method, question-answering (Q&A) technology has become a core technical support for addressing current issues such as low approval efficiency and difficulty in obtaining information. However, existing Q&A technologies suffer from significant hallucination problems and limitations in considering spatiotemporal factors in the land use approval domain. To effectively address these issues, this study proposes a spatiotemporal–semantic coupling intelligent Q&A method for land use approval based on knowledge graphs (KGs) and intelligent agent technology, aiming to enhance the efficiency and quality of land use approval. Firstly, a land use approval knowledge graph (LUAKG) is constructed, systematically integrating domain knowledge such as policy clauses, legal regulations, and approval procedures. Then, by combining large language models (LLMs) and intelligent agent technology, a spatiotemporal–semantic coupling Q&A framework is designed. Through the use of spatiotemporal analysis tools, this framework can comprehensively consider spatial, temporal, and semantic factors when handling land approval tasks, enabling dynamic decision-making and precise reasoning. The research results show that, compared to traditional Q&A based on LLMs and Q&A based on retrieval-enhanced generation (RAG), the proposed method improves accuracy by 16% and 9% in general knowledge Q&A tasks. In the project review Q&A task, F1 scores and accuracy increase by 2% and 9%, respectively, compared to RAG-QA. Particularly, under the spatiotemporal–semantic multidimensional analysis, the improvement in F1 score and accuracy ranges from 2 to 6% and 7 to 10%, respectively. Full article

Predictive maintenance (PdM) represents a significant evolution in maintenance strategies. However, challenges such as system integration complexity, data quality, and data availability are intricately intertwined, collectively impacting the successful deployment of PdM systems. Recently, large model-based agents, or agentic artificial intelligence (AI), have evolved from simple task automation to active problem-solving and strategic decision-making. As such, we propose an AI agent-enabled PdM method that leverages an agentic AI development platform to streamline the development of a multimodal data-based fault detection agent, a RAG (retrieval-augmented generation)-based fault classification agent, a large model-based fault diagnosis agent, and a digital twin-based fault handling simulation agent. This approach breaks through the limitations of traditional PdM, which relies heavily on single models. This combination of “AI workflow + large reasoning models + operational knowledge base + digital twin” integrates the concepts of BaaS (backend as a service) and LLMOps (large language model operations), constructing an end-to-end intelligent closed loop from data perception to decision execution. Furthermore, a tentative prototype is demonstrated to show the technology stack and the system integration methods of the agentic AI-based PdM. Full article

Smart contracts have become integral to decentralized applications, yet their programmability introduces critical security risks, exemplified by high-profile exploits such as the DAO and Parity Wallet incidents. Existing vulnerability detection methods, including static and dynamic analysis, as well as machine learning-based approaches, often struggle with emerging threats and rely heavily on large, labeled datasets. This study investigates the effectiveness of open-source, lightweight large language models (LLMs) fine-tuned using parameter-efficient techniques, including Quantized Low-Rank Adaptation (QLoRA), for smart contract vulnerability detection. We introduce the EVuLLM dataset to address the scarcity of diverse evaluation resources and demonstrate that our fine-tuned models achieve up to 94.78% accuracy, surpassing the performance of larger proprietary models, while significantly reducing computational requirements. Moreover, we emphasize the advantages of lightweight models deployable on local hardware, such as enhanced data privacy, reduced reliance on internet connectivity, lower infrastructure costs, and improved control over model behavior, factors that are especially critical in security-sensitive blockchain applications. We also explore Retrieval-Augmented Generation (RAG) as a complementary strategy, achieving competitive results with minimal training. Our findings highlight the practicality of using locally hosted LLMs for secure, efficient, and reproducible smart contract analysis, paving the way for broader adoption of AI-driven security in blockchain ecosystems. Full article

In this paper, an image-driven regional pansharpening technique based on simplex optimization analysis with a graph-based superpixel segmentation strategy is proposed. This fusion approach optimally combines spatial information derived from a high-resolution panchromatic (PAN) image and spectral information captured from a low-resolution multispectral (MS) image to generate a unique comprehensive high-resolution MS image. As the performance of such a fusion method relies on the choice of the fusion strategy, and in particular, on the way the algorithm is used for estimating gain coefficients, our proposal is dedicated to computing the injection gains over a graph-driven segmentation map. The graph-based segments are obtained by applying simple linear iterative clustering (SLIC) on the MS image followed by a region adjacency graph (RAG) merging stage. This graphical representation of the segmentation map is used as guidance for spatial information to be injected during fusion processing. The high-resolution MS image is achieved by inferring locally the details in accordance with the local simplex injection fusion rule. The quality improvements achievable by our proposal are evaluated and validated at reduced and at full scales using two high resolution datasets collected by GeoEye-1 and WorldView-3 sensors. Full article

This study presents TeS v.3, a thermodynamic equilibrium simulator integrated with an artificial intelligence agent (AI), ThermoAgent, to enhance the analysis of complex chemical systems. Developed in Python, the simulator employs Gibbs energy minimization for isothermal reactors and entropy maximization for adiabatic reactors. ThermoAgent leverages the LangChain framework to interpret natural language commands, autonomously execute simulations, and query a scientific knowledge base through a Retrieval-Augmented Generation (RAG) approach. The validation of TeS v.3 demonstrated high accuracy, with coefficients of determination (R² > 0.95) compared to reference simulation data and strong correlation (R² > 0.88) with experimental data from the steam methane reforming (SMR) process. The SMR analysis correctly distinguished the high conversions in isothermal reactors from the limited conversions in adiabatic reactors, due to the reaction temperature drop. ThermoAgent successfully executed simulations and provided justified analyses, combining generated data with information from reference publications. The successful integration of the simulator with the AI agent represents a significant advancement, offering a powerful tool that accurately calculates equilibrium and accelerates knowledge extraction through intuitive interaction. Full article

Cyberspace search engines (CSEs) are systems designed to search and index information about cyberspace assets. Effectively mining data across diverse platforms is hindered by the complexity and diversity of different CSE syntaxes. While Text-to-CSEQL offers a promising solution by translating natural language (NL) questions into cyberspace search engine query language (CSEQL), existing prompt-based methods still struggle due to the platform-specific intricacies of CSEQL. To address this limitation, we propose an LLM-based approach leveraging Retrieval-Augmented Generation (RAG). Specifically, to overcome the inability of traditional methods to retrieve relevant syntax fields effectively, we propose a novel hybrid retrieval mechanism combining keyword and dense retrieval, leveraging both field values and their semantic descriptions. Furthermore, we integrate these retrieved fields and the relevant few-shot examples into a redesigned prompt template adapted from the COSTAR framework. For comprehensive evaluation, we construct a Text-to-CSEQL dataset and introduce a new domain-specific metric, field match (FM). Extensive experiments demonstrate our method’s ability to adapt to platform-specific characteristics. Compared to prompt-based methods, it achieves an average accuracy improvement of 43.15% when generating CSEQL queries for diverse platforms. Moreover, our method also outperforms techniques designed for single-platform CSEQL generation. Full article

Large Language Models (LLMs) are reshaping recommendation systems through enhanced language understanding, reasoning, and integration with structured data. This systematic review analyzes 88 studies published between 2023 and 2025, categorized into three thematic areas: data processing, technical identification, and LLM-based recommendation architectures. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, the review highlights key trends such as the use of knowledge graphs, Retrieval-Augmented Generation (RAG), domain-specific fine-tuning, and robustness improvements. Findings reveal that while LLMs significantly advance semantic reasoning and personalization, challenges remain in hallucination mitigation, fairness, and domain adaptation. Technical innovations, including graph-augmented retrieval methods and human-in-the-loop validation, show promise in addressing these limitations. The review also considers the broader macroeconomic implications associated with the deployment of LLM-based systems, particularly as they relate to scalability, labor dynamics, and resource-intensive implementation in real-world recommendation contexts, emphasizing both productivity gains and potential labor market shifts. This work provides a structured overview of current methods and outlines future directions for developing reliable and efficient LLM-based recommendation systems. Full article

Floods pose a growing threat globally, causing tragic loss of life, billions in economic damage annually, and disproportionately affecting socio-economically vulnerable populations. This paper aims to improve flood-risk communication for community leaders by exploring the application of artificial intelligence. We categorize U.S. flood information sources, review communication modalities and channels, synthesize the literature on community leaders’ roles in risk communication, and analyze existing technological tools. Our analysis reveals three key challenges: the fragmentation of flood information, information overload that impedes decision-making, and the absence of a unified communication platform to address these issues. We find that AI techniques can organize data and significantly enhance communication effectiveness, particularly when delivered through infographics and social media channels. Based on these findings, we propose FLAI (Flood Language AI), an AI-driven flood communication platform that unifies fragmented flood data sources. FLAI employs knowledge graphs to structure fragmented data sources and utilizes a retrieval-augmented generation (RAG) framework to enable large language models (LLMs) to produce contextualized narratives, including infographics, maps, and cost–benefit analyses. Beyond flood management, FLAI’s framework demonstrates how AI can transform public service data management and institutional AI readiness. By centralizing and organizing information, FLAI can significantly reduce the cognitive burden on community leaders, helping them communicate timely, actionable insights to save lives and build flood resilience. Full article

Recent advancements in large language models (LLMs) have significantly enhanced the effectiveness of Retrieval-Augmented Generation (RAG) systems. This study focuses on the development and evaluation of a domain-specific AI chatbot designed to support international student admissions by leveraging LLM-based RAG pipelines. We implement and compare multiple pipeline configurations, combining retrieval methods (e.g., Dense, MMR, Hybrid), chunking strategies (e.g., Semantic, Recursive), and both open-source and commercial LLMs. Dual evaluation datasets of LLM-generated and human-tagged QA sets are used to measure answer relevancy, faithfulness, context precision, and recall, alongside heuristic NLP metrics. Furthermore, latency analysis across different RAG stages is conducted to assess deployment feasibility in real-world educational environments. Results show that well-optimized open-source RAG pipelines can offer comparable performance to GPT-4o while maintaining scalability and cost-efficiency. These findings suggest that the proposed chatbot system can provide a practical and technically sound solution for international student services in resource-constrained academic institutions. Full article

Large language models (LLMs) often tend to hallucinate, especially in domain-specific tasks and tasks that require reasoning. Previously, we introduced SubGraph Retrieval Augmented Generation (SG-RAG) as a novel Graph RAG method for multi-hop question answering. SG-RAG leverages Cypher queries to search a given knowledge graph and retrieve the subgraph necessary to answer the question. The results from our previous work showed the higher performance of our method compared to the traditional Retrieval Augmented Generation (RAG). In this work, we further enhanced SG-RAG by proposing an additional step called Merging and Ordering Triplets (MOT). The new MOT step seeks to decrease the redundancy in the retrieved triplets by applying hierarchical merging to the retrieved subgraphs. Moreover, it provides an ordering among the triplets using the Breadth-First Search (BFS) traversal algorithm. We conducted experiments on the MetaQA benchmark, which was proposed for multi-hop question-answering in the movies domain. Our experiments showed that SG-RAG MOT provided more accurate answers than Chain-of-Thought and Graph Chain-of-Thought. We also found that merging (up to a certain point) highly overlapping subgraphs and defining an order among the triplets helped the LLM to generate more precise answers. Full article

The exponential increase in textual data generated across sectors such as healthcare, finance, and smart manufacturing has intensified the need for effective Big Data analytics. Large language models (LLMs) have become critical tools because of their advanced language processing capabilities. However, their static nature limits their ability to incorporate real-time and domain-specific knowledge. Retrieval-augmented generation (RAG) addresses these limitations by enriching LLM outputs through external content retrieval. Nevertheless, traditional RAG systems remain inefficient, often exhibiting high retrieval latency, redundancy, and diminished response quality when scaled to large datasets. This paper proposes an innovative structured RAG framework specifically designed for large-scale Big Data analytics. The framework transforms unstructured partial prompts into structured semantically coherent partial prompts, leveraging element-specific embedding models and dimensionality reduction techniques, such as principal component analysis. To further improve the retrieval accuracy and computational efficiency, we introduce a multi-level filtering approach integrating semantic constraints and redundancy elimination. In the experiments, the proposed method was compared with structured-format RAG. After generating prompts utilizing two methods, silhouette scores were computed to assess the quality of embedding clusters. The proposed method outperformed the baseline by improving the clustering quality by 32.3%. These results demonstrate the effectiveness of the framework in enhancing LLMs for accurate, diverse, and efficient decision-making in complex Big Data environments. Full article

The increasing complexity and scale of electronic health records (EHRs) demand advanced tools for efficient data retrieval, summarization, and comparative analysis in clinical practice. MERA (Medical Electronic Records Assistant) is a Retrieval-Augmented Generation (RAG)-based AI system that addresses these needs by integrating domain-specific retrieval with large language models (LLMs) to deliver robust question answering, similarity search, and report summarization functionalities. MERA is designed to overcome key limitations of conventional LLMs in healthcare, such as hallucinations, outdated knowledge, and limited explainability. To ensure both privacy compliance and model robustness, we constructed a large synthetic dataset using state-of-the-art LLMs, including Mistral v0.3, Qwen 2.5, and Llama 3, and further validated MERA on de-identified real-world EHRs from the MIMIC-IV-Note dataset. Comprehensive evaluation demonstrates MERA’s high accuracy in medical question answering (correctness: 0.91; relevance: 0.98; groundedness: 0.89; retrieval relevance: 0.92), strong summarization performance (ROUGE-1 F1-score: 0.70; Jaccard similarity: 0.73), and effective similarity search (METEOR: 0.7–1.0 across diagnoses), with consistent results on real EHRs. The similarity search module empowers clinicians to efficiently identify and compare analogous patient cases, supporting differential diagnosis and personalized treatment planning. By generating concise, contextually relevant, and explainable insights, MERA reduces clinician workload and enhances decision-making. To our knowledge, this is the first system to integrate clinical question answering, summarization, and similarity search within a unified RAG-based framework. Full article