Intelligent Document Query System using Retrieval-Augmented Generation (RAG)

Time series (TS) data are ubiquitous across various application areas, rendering time series forecasting (TSF) a fundamental task. With the astounding advances in large language models (LLMs), a variety of methods have been developed to adapt LLMs for time series forecasting. Despite unlocking the potential of LLMs in comprehending TS data, existing methods are inherently constrained by their shallow integration of TS information, wherein LLMs typically access TS representations at shallow layers, primarily at the input layer. This causes the influence of TS representations to progressively fade in deeper layers and eventually leads to ineffective adaptation between textual embeddings and TS representations. In this paper, we propose the Multi-layer Steerable Embedding Fusion (MSEF), a novel framework that enables LLMs to directly access time series patterns at all depths, thereby mitigating the progressive loss of TS information in deeper layers. Specifically, MSEF leverages off-the-shelf time series foundation models to extract semantically rich embeddings, which are fused with intermediate text representations across LLM layers via layer-specific steering vectors. These steering vectors are designed to continuously optimize the alignment between time series and textual modalities and facilitate a layer-specific adaptation mechanism that ensures efficient few-shot learning capabilities. Experimental results on seven benchmarks demonstrate significant performance improvements by MSEF compared with baselines, with an average reduction of 31.8% in terms of MSE. The code is available at https://github.com/One1sAll/MSEF.

In order to minimize the impact of influenza on public health, accurate early forecasting is essential. Various deep-learning-based models have been proposed to predict future influenza occurrences by capturing temporal/regional patterns from past occurrence time-series data. However, the prediction performance of these unimodal approaches is limited because they extract knowledge only from collected data, and users cannot input contextual information and domain knowledge to them. Recently, large language models (LLMs) have demonstrated the potential to improve prediction accuracy by linking contextual text information to time-series predictions. In this paper, we propose MIFlu, a multimodal influenza forecasting scheme that can fuse contextual text information to time-series influenza occurrences using two LLMs. It first extracts text embeddings from the user's text prompts that contain contextual information using a text-embedding LLM. Then, MIFlu fuses the text embeddings and time-series embeddings and uses the fused embeddings to predict future occurrences using a forecasting LLM. In extensive experiments using public national/regional influenza datasets, MIFlu outperforms other predictive models, improving prediction performance by up to 26.2% compared to state-of-the-art models. We also analyze the effect of various textual input embedders, hyperparameters, and the amount of training data on forecasting accuracy.

Legacy software, particularly from the 1990s, often becomes obsolete due to aging hardware and outdated software environments. Traditionally, software modernization required extensive manual effort, involving reverse engineering, code rewriting, and re-architecting. However, advancements in large language models (LLMs) have introduced new possibilities for automating software translation and modernization. This paper explores the feasibility of using LLMs for modernizing 90s-era Windows applications, specifically migrating legacy C and C[Formula: see text] code to Python. Our methodology includes decompilation, source code analysis, automated translation using ChatGPT, and user interface reconstruction. We empirically evaluate three software projects by analyzing LLM-based translation accuracy across different code structures, including algorithmic logic, file handling, and graphical interfaces. Results indicate that while LLMs achieve high translation accuracy [Formula: see text] for structured code, challenges persist in handling decompiled code and user interface generation. The study provides insights into the effectiveness and limitations of LLMs in real-world software renovation, offering guidelines for leveraging machine learning in legacy system modernization. These findings contribute to both academic research and practical applications, suggesting a pathway for cost-effective and scalable legacy software migration.

Extracting information from unstructured and semi-structured geoscience literature is a crucial step in conducting geological research. The traditional machine learning extraction paradigm requires a substantial amount of high-quality manually annotated data for model training, which is time-consuming, labor-intensive, and not easily transferable to new fields. Recently, large language models (LLMs) (e.g., ChatGPT, GPT-4, and LLaMA), have shown great performance in various natural language processing (NLP) tasks, such as question answering, machine translation, and text generation. A substantial body of work has demonstrated that LLMs possess strong in-context learning (ICL) and even zero-shot learning capabilities to solve downstream tasks without specifically designed supervised fine-tuning. In this paper, we propose utilizing LLMs for geoscience literature information extraction. Specifically, we design a hierarchical PDF parsing pipeline and an automated knowledge extraction process, which can significantly reduce the need for manual data annotation, assisting geoscientists in literature data mining. For the hierarchical PDF parsing pipeline, firstly, a document layout detection model fine-tuned on geoscience literature is employed for layout detection, obtaining layout detection information for the document. Secondly, based on the document layout information, an optical character content parsing model is used for content parsing, obtaining the text structure and plain text corresponding to the content. Finally, the text structure and plain text are combined and reconstructed to ultimately obtain the parsed structured data. For the automated knowledge extraction process, firstly, the parsed long text is segmented into paragraphs to adapt to the input length limit of LLMs. Subsequently, a few-shot prompting method is employed for structured knowledge extraction, encompassing two tasks: attribute value extraction and triplet extraction. In attribute value extraction, prompts are generated automatically by the LLMs based on the subdomain and attribute names, facilitating the location and extraction of values related to subdomain attribute names in the text. For triplet extraction, the LLMs employ a procedural approach to entity extraction, entity type extraction, and relation extraction, following the knowledge graph structure pattern. Finally, the extracted structured knowledge is stored in the form of knowledge graphs, facilitating further analysis and integration of various types of knowledge from the literature. Our proposed approach turns out to be simple, flexible, and highly effective in geoscience literature information extraction. Demonstrations of information extraction in subdomains such as radiolarian fossils and fluvial facies have yielded satisfactory results. The extraction efficiency has significantly improved, and feedback from domain experts indicates a relatively high level of accuracy in the extraction process. The extracted results can be used to construct a foundational knowledge graph for geoscience literature, supporting the comprehensive construction and efficient application of a geoscience knowledge graph.

Large language models (LLMs) offer significant promise as a knowledge source for task learning. Prompt engineering has been shown to be effective for eliciting knowledge from an LLM, but alone it is insufficient for acquiring relevant, situationally grounded knowledge for an embodied agent learning novel tasks. We describe a cognitive-agent approach, STARS, that extends and complements prompt engineering, mitigating its limitations and thus enabling an agent to acquire new task knowledge matched to its native language capabilities, embodiment, environment, and user preferences. The STARS approach is to increase the response space of LLMs and deploy general strategies, embedded within the autonomous agent, to evaluate, repair, and select among candidate responses produced by the LLM. We describe the approach and experiments that show how an agent, by retrieving and evaluating a breadth of responses from the LLM, can achieve 77-94% task completion in one-shot learning without user oversight. The approach achieves 100% task completion when human oversight (such as an indication of preference) is provided. Further, the type of oversight largely shifts from explicit, natural language instruction to simple confirmation/discomfirmation of high-quality responses that have been vetted by the agent before presentation to a user.

Text clustering is an important method for organising the increasing volume of digital content, aiding in the structuring and discovery of hidden patterns in uncategorised data. The effectiveness of text clustering largely depends on the selection of textual embeddings and clustering algorithms. This study argues that recent advancements in large language models (LLMs) have the potential to enhance this task. The research investigates how different textual embeddings, particularly those utilised in LLMs, and various clustering algorithms influence the clustering of text datasets. A series of experiments were conducted to evaluate the impact of embeddings on clustering results, the role of dimensionality reduction through summarisation, and the adjustment of model size. The findings indicate that LLM embeddings are superior at capturing subtleties in structured language. OpenAI’s GPT-3.5 Turbo model yields better results in three out of five clustering metrics across most tested datasets. Most LLM embeddings show improvements in cluster purity and provide a more informative silhouette score, reflecting a refined structural understanding of text data compared to traditional methods. Among the more lightweight models, BERT demonstrates leading performance. Additionally, it was observed that increasing model dimensionality and employing summarisation techniques do not consistently enhance clustering efficiency, suggesting that these strategies require careful consideration for practical application. These results highlight a complex balance between the need for refined text representation and computational feasibility in text clustering applications. This study extends traditional text clustering frameworks by integrating embeddings from LLMs, offering improved methodologies and suggesting new avenues for future research in various types of textual analysis.

This article sets in with the question whether current or foreseeable transformer-based large language models (LLMs), such as the ones powering OpenAI’s ChatGPT, could be language users in a way comparable to humans. It answers the question negatively, presenting the following argument. Apart from niche uses, to use language means to act. But LLMs are unable to act because they lack intentions. This, in turn, is because they are the wrong kind of being: agents with intentions need to be autonomous organisms while LLMs are heteronomous mechanisms. To conclude, the article argues, based on structural aspects of transformer-based LLMs, that these LLMs have taken a first step away from mechanistic artificiality to autonomous self-constitution, which means that these models are (slowly) moving into a direction that someday might result in non-human, but equally non-artificial agents, thus subverting the time-honored Kantian distinction between organism and mechanism.

In this work, a thorough mathematical framework for incorporating Large Language Models (LLMs) into gamified systems is presented with an emphasis on improving task dynamics increasing user engagement, and improving reward systems. Personalized feedback adaptive learning and dynamic content creation are all made possible by the integration of LLMs and are crucial for improving user engagement and system performance. A simulated environment is used to test the framework's adaptability and demonstrate its potential for real-world applications in a variety of industries including business healthcare and education. The findings demonstrate how LLMs can offer customized experiences that raise system effectiveness and user retention. This study also examines the difficulties this framework aims to solve highlighting its importance in maximizing involvement and encouraging sustained behavioral change in a range of sectors.

Background: Virtual coaching can help people adopt new healthful behaviors by encouraging them to set specific goals and helping them review their progress. One challenge in creating such systems is analyzing clients’ statements about their activities. Limiting people to selecting among predefined answers detracts from the naturalness of conversations and user engagement. Large Language Models (LLMs) offer the promise of covering a wide range of expressions. However, using an LLM for simple entity extraction would not necessarily perform better than functions coded in a programming language, while creating higher long-term costs. Methods: This study uses a real data set of annotated human coaching dialogs to develop LLM-based models for two training scenarios: one that generates pattern-matching functions and the other which does direct extraction. We use models of different sizes and complexity, including Meta-Llama, Gemma, and ChatGPT, and calculate their speed and accuracy. Results: LLM-generated pattern-matching functions took an average of 10 milliseconds (ms) per item as compared to 900 ms. (ChatGPT 3.5 Turbo) to 5 s (Llama 2 70B). The accuracy for pattern matching was 99% on real data, while LLM accuracy ranged from 90% (Llama 2 70B) to 100% (ChatGPT 3.5 Turbo), on both real and synthetically generated examples created for fine-tuning. Conclusions: These findings suggest promising directions for future research that combines both methods (reserving the LLM for cases that cannot be matched directly) or that use LLMs to generate synthetic training data with more expressive variety which can be used to improve the coverage of either generated codes or fine-tuned models.

Burnout has become a significant concern in high-stress industries, impacting mental health and productivity. Traditional burnout management solutions lack real-time adaptability and personalized intervention strategies. This research explores an AI-driven burnout management mechanism leveraging Generative AI, MongoDB Atlas, Python, and Large Language Models (LLMs) to provide real-time detection, personalized coping strategies, and predictive analytics. The system collects and stores user responses, behavioral patterns, and sentiment data in MongoDB Atlas, while LLMs analyze and generate tailored burnout mitigation recommendations. Experimental results demonstrate the effectiveness of the proposed approach in early burnout detection, personalized intervention, and user engagement, showing improvements in stress reduction and work-life balance. The proof of concept for AI-powered burnout detection and intervention demonstrates the technical feasibility of using Large Language Models (LLMs), Generative AI, MongoDB Atlas, and Python-based automation for real-time mental health support. The system successfully classifies burnout severity using LLMs trained on sentiment analysis, achieving an accuracy of 91.2%, which significantly outperforms traditional rule-based sentiment models. The use of MongoDB Atlas for scalable data storage ensures real-time burnout tracking, while FastAPI-based backend automation allows for seamless integration across platforms. Furthermore, Generative AI provides personalized stress interventions, dynamically adapting to user feedback and improving engagement. These technical validations highlight how AI-driven burnout management can be scalable, real-time, and enterprise-ready, paving the way for AI-powered diagnostics in mental health applications across various industries.

This study investigated the feasibility of using large language models (LLMs) to infer problematic Instagram use, which refers to excessive or compulsive engagement with the platform that negatively impacts users’ daily functioning, productivity, or well-being, from a limited set of metrics of user engagement in the platform. Specifically, we explored whether OpenAI’s GPT-4o and Google’s Gemini 1.5 Pro could accurately predict self-reported problematic use tendencies based solely on readily available user engagement metrics like daily time spent on the platform, weekly posts and stories, and follower/following counts. Our sample comprised 775 Italian Instagram users (61.6% female; aged 18–63), who were recruited through a snowball sampling method. Item-level and total scores derived by querying the LLMs’ application programming interfaces were correlated with self-report items and the total score measured via an adapted Bergen Social Media Addiction Scale. LLM-inferred scores showed positive correlations with both item-level and total scores for problematic Instagram use. The strongest correlations were observed for the total scores, with GPT-4o achieving a correlation of r = 0.414 and Gemini 1.5 Pro achieving a correlation of r = 0.319. In cross-validated regression analyses, adding LLM-generated scores, especially from GPT-4o, significantly improved the prediction of problematic Instagram use compared to using usage metrics alone. GPT-4o’s performance in random forest models was comparable to models trained directly on Instagram metrics, demonstrating its ability to capture complex, non-linear relationships indicative of addiction without needing extensive model training. This study provides compelling preliminary evidence for the use of LLMs in inferring problematic Instagram use from limited data points, opening exciting new avenues for research and intervention.

In this paper, we introduce a cutting-edge solution to the complex challenge of managing operational costs in unconventional asset development, particularly concerning continuous well drilling, completion, and field maintenance operations within the oil and gas sector. These complicated field operations involve hundreds of service providers and incur vast volumes of invoices every year. Tremendous values are lost due to missed opportunities to improve contracting strategy, optimize material and equipment supplies, and identify cost-prohibitive design elements during drilling, completion, production, and plant maintenance through proper spend categorization. Leveraging the power of a machine learning solution, Large Language Model (LLM), and an interactive user interface, we automate the challenging task of categorizing millions of invoices from diverse service providers. Techniques including sentence transformation embedding, transfer learning, fine-tuning, and re-training process are employed to enhance model performance and adaptability to diverse invoice types. Through rigorous model training and iterative refinement facilitated by the user interface, our approach attains an impressive accuracy exceeding 90% across all regions. Results prove that the Large Language Model has a wide application in business optimization during unconventional asset development. This automated algorithm provides real-time insights through spending and significantly reduces the turnaround time. This study also enables direct identification of cost-saving opportunities such as potential reductions in fuel expenses related to drilling and completion activities. The advanced analytics capabilities following the modeling effort allow engineers and analysts in multiple functions to identify cost-saving opportunities through customizing visualization tools in various areas. This paper presents original methodology to adopt the Artificial Intelligence, i.e., Large Language Model, in the oil and gas industry and demonstrate its values with case histories.

Large language models (LLMs) are increasingly used by retail traders to interpret information and design complex strategies, yet existing adoption constructs do not capture the decision-time experience of being cognitively scaffolded by an LLM. We define Perceived Cognitive Assistance (PCA) as the trader’s felt expansion of cognitive capability at the moment of trading decision when an LLM is available, and we report initial content validation of a PCA item pool. Study 1 specified the PCA content domain using a two-tier qualitative corpus (8 interviews and 44 YouTube narratives on LLM-assisted trading, plus 24 qualitative and mixed-method studies on robo-advice and social trading). Reflexive thematic analysis yielded five facilitative assistance facets and one adjacent risk facet (over-reliance), and these were translated into a 16-item PCA pool. Study 2 used a naïve-judge sort-and-rate task with 48 retail traders to test whether items show definitional correspondence to PCA and definitional distinctiveness from similar constructs: perceived usefulness, perceived ease of use, trust in the LLM, and trading self-efficacy. The resulting 9 item set is ready for subsequent factor-analytic and predictive validation. This study advances our understanding of how large language models shape retail trading behaviour by identifying and empirically grounding Perceived Cognitive Assistance as the decision-time psychological experience through which LLMs cognitively scaffold traders, clarifying how LLM use differs from generic technology adoption, trust, or self-efficacy effects.

DocuMind AI is an advanced document-based question answering system developed using Retrieval-Augmented Generation (RAG) to address challenges in extracting meaningful information from unstructured data. Organizations often struggle to extract useful information from large volumes of unstructured documents. Traditional keyword-based search methods often fail to capture semantic context, resulting in irrelevant outputs. To overcome this limitation, the proposed system integrates semantic search with large language models (LLMs) to generate accurate and context-aware responses. The system utilizes vector embeddings to represent document content and retrieve relevant information efficiently. It is implemented using a Java Spring Boot backend integrated with LangChain4j, which manages document ingestion, text chunking, embedding generation, and retrieval processes. Users interact with the system through a React.js-based interface, allowing them to upload documents and query in natural language. The system supports multiple document formats such as PDF and DOCX, ensuring flexibility in real-world applications. Additionally, JWTbased authentication is implemented to provide secure access to the system. The modular and scalable architecture makes DocuMind suitable for enterprise-level deployment, bridging the gap between traditional information retrieval systems and modern AI-driven solutions. Keywords: Retrieval-Augmented Generation, Natural Language Processing, Large Language Models, Vector Database, Java Spring Boot, LangChain4j, PostgreSQL pgvector, Cosine Similarity.

Accurate color alignment in text-to-image (T2I) generation is critical for applications such as fashion, product visualization, and interior design, yet current diffusion models struggle with nuanced and compound color terms (e.g., Tiffany blue, baby pink), often producing images that are misaligned with human intent. Existing approaches rely on cross-attention manipulation, reference images, or fine-tuning but fail to systematically resolve ambiguous color descriptions. To precisely render colors under prompt ambiguity, we propose a training-free framework that enhances color fidelity by leveraging a large language model (LLM) to disambiguate color-related prompts and guiding color blending operations directly in the text embedding space. Our method first employs a large language model (LLM) to resolve ambiguous color terms in the text prompt, and then refines the text embeddings based on the spatial relationships of the resulting color terms in the CIELab color space. Unlike prior methods, our approach improves color accuracy without requiring additional training or external reference images. Experimental results demonstrate that our framework improves color alignment without compromising image quality, bridging the gap between text semantics and visual generation. All supplementary materials are available at https://Sung-Lin.github.io/TintBench/.