Search Results (318)

In artificial breeding of forest musk deer (Moschus berezovskii), common diseases such as abscess, enteritis, pneumonia, and parasitic infections exhibit persistently high morbidity rates. The early symptoms of certain diseases are often insidious and difficult to discern. Conventional manual inspection routines not only fail to achieve accurate diagnosis but also frequently disturb the animals, induce stress responses, and consequently delay optimal treatment windows. To address this practical challenge, this study employs an improved BRW-GPLinker joint entity-relationship extraction approach to perform integrated extraction and structural organization of disease entities, symptom manifestations, etiological associations, and preventive and therapeutic measures from farming literature and clinical records, thereby constructing a disease knowledge graph for forest musk deer. Through the introduction of a Boundary-Aware Module for refined entity boundary detection, a Relative Distance Bias Module to mitigate pairing errors in dense contexts, and a Weighted Sparse Multi-label Cross-Entropy loss function to enhance recall for infrequent relations, the proposed model achieves an F1 score of 0.887 on a self-constructed dataset and demonstrates favorable generalization capability on medical-domain datasets. By transforming fragmented clinical logs and manuals into structured medical associations, this knowledge graph facilitates rapid retrieval of forest musk deer disease information, thereby enhancing veterinary decision-making efficiency and assisting forest musk deer health management. Full article

Reasoning-intensive multimodal retrieval suffers from a counter-intuitive bottleneck: on MM-BRIGHT multimodal-to-text (Query+Image → Documents), the strongest dense multimodal encoder reaches only 27.6 nDCG@10 and the rest of the dense vision–language retrievers cluster between 10.0 and 23.0. The visual signal, encoded as a dense vector, adds noise rather than evidence; even augmenting strong text retrievers with raw image captions degrades performance by up to 12.0 points. We propose VISA, a Visual Symbolic Agent that re-casts multimodal-to-text as text retrieval over three parallel streams. A Vision LLM is dispatched in three roles via separate prompts: a zero-shot router that classifies the query image into up to three parser types from a fixed taxonomy of nine (chart, circuit, equation, screenshot, code, figure, diagram, map, photograph); typed parsers that extract structured text per type; and a holistic captioner. The agent constructs three text streams (raw query, query ⊕ symbolic, query ⊕ caption), scores each with a single frozen 4B-parameter retrieval LLM, and fuses the per-document scores via Reciprocal Rank Fusion or a confidence-weighted linear combination. The whole agent contains no trainable parameters. The key novelty is a change of substrate: rather than projecting the query image into a dense multimodal vector that competes with text, VISA is, to our knowledge, the first retrieval system to convert the image into typed symbolic text and keep retrieval entirely text-side, so that a frozen text retriever can match the literal tokens (axis values, variable names, function signatures) that answering documents actually contain. Across all 29 MM-BRIGHT multimodal-to-text domains, VISA achieves 32.4 nDCG@10, an absolute improvement of +4.8 over the strongest dense multimodal encoder and substantially larger margins over the remaining six dense vision–language baselines. Per-domain analysis shows VISA maintains its margin across STEM and software domains where image content is structure-heavy. In practical terms, VISA is training-free and model-agnostic: it requires no fine-tuning, reuses any off-the-shelf vision LLM and text retriever, caches all per-image parsing so re-runs cost only three query encodes, and can therefore be dropped into an existing text-retrieval stack to add reasoning-intensive multimodal capability without building or training a multimodal encoder. Full article

Industrial robot after-sales question answering often depends on multimodal evidence, such as error screenshots, interface displays, and wiring diagrams, which are difficult for conventional text-based retrieval-augmented generation (RAG) systems to exploit effectively. To address this issue, we design a dual-channel multimodal RAG system that converts image content into retrievable textual knowledge through the collaboration of optical character recognition (OCR) and structured semantic description. In the proposed system, OCR is used to extract explicit textual cues, such as error codes, parameter fields, and interface prompts, while expert-authored semantic descriptions complement implicit visual evidence, including device parts, fault phenomena, and contextual scene information. The transformed knowledge is further integrated into a hybrid retrieval pipeline that combines dense retrieval and BM25, followed by Reciprocal Rank Fusion (RRF) and Maximal Marginal Relevance (MMR) reordering to improve both relevance and contextual diversity. Experiments on a real-world industrial robot after-sales dataset show that the proposed method achieves an overall question-answering accuracy of 87.9%, outperforming the LLM-only baseline by 35.6 percentage points. For image-related questions, accuracy improves from 46.7% to 83.3%. These results indicate that the proposed framework provides a deployment-friendly and interpretable system-level alternative to end-to-end multimodal model fine-tuning for industrial after-sales question answering. Full article

A Retrieval-Augmented Generation (RAG) approach is widely used as a key element for intelligent assistants. However, the knowledge extraction stage from technical text corpora is fraught with difficulties due to the presence of highly specialized terminology, tables, and abbreviations. The goal of this study is to develop methodological support for knowledge extraction for an intelligent assistant for a technical specialist in the field of microcontroller-based device design. This study systematically compares and analyzes the computational performance of knowledge extraction methods and their various combinations. The results showed that the hybrid version of the baseline methods (hybrid_v2_dense) provides the best R@1 (45.2%), MRR@5 (49.8%) and nDCG@5 (52.0%) values, while the R@5 level remains comparable to BM25. Among the extended configurations of the hybrid_v2 family, the best R@5 value (57.7%) is achieved by the hybrid_v2_dense_splade method, while the best values of R@1 (48.9%), MRR@5 (52.1%), and nDCG@5 (53.7%) are achieved by the hybrid_v2_dense_unicoil method. Based on the obtained results, an expert decision tree was formed for selecting the knowledge extraction module configuration considering hardware limitations. These results provide experimental evidence of the effectiveness of the developed methodological support for knowledge extraction for an intelligent assistant of a technical specialist. Full article

This study addresses the growing need to convert unstructured legal and policy documents into formal computational models that support transparent decision-making. The purpose of the work is to develop an applied framework that connects Legal NLP and policy information extraction with canonical combinatorial decision models, including set cover, set packing, subset sum, vertex cover, and independent set. The proposed method combines MPNet-based dense semantic retrieval for locating relevant legal passages, a Legal NLP layer for extracting obligations, prohibitions, exceptions, thresholds, and eligibility conditions, and a formal modeling stage that maps the extracted constraints to NP-complete formulations, including set cover, set packing, subset sum, vertex cover, and independent set. The framework is designed to transform regulatory text into machine-interpretable structures suitable for constraint-aware reasoning and policy analysis. The results show that the integration of semantic retrieval and structured legal information extraction improves the consistency, interpretability, and practical usability of formal problem construction from legal and policy documents. The proposed approach provides a reproducible bridge between legal text analytics and combinatorial decision modeling and supports legal decision support, compliance analysis, and policy-oriented intelligent systems. Full article

Large language models (LLMs) are increasingly being used in radiology-related workflows, but their application to reference, regulatory, and methodological queries remains limited by hallucinations and the static nature of model knowledge. This study aimed to develop and evaluate a retrieval-augmented generation (RAG) system for radiologists designed to provide grounded responses to such queries. A knowledge base was created through a survey of practicing radiologists and expert validation of sources, resulting in a corpus of 1049 documents. The system incorporated structured document parsing, a two-level parent–child vector database, hybrid dense–sparse retrieval, reranking, and a local large language model. Performance was assessed through functional testing, automated LLM-as-a-judge metrics, and multireader expert evaluation by 16 radiologists using 400 technical queries. No hallucinations were detected in the 77-query functional testing set during expert review. On the full technical dataset, automated Contextual Precision, Contextual Recall, and Answer Relevancy were 0.735, 0.881, and 0.890, respectively. Expert evaluation showed high response accuracy (mean, 4.53/5) and high expert-assessed Contextual Precision (0.886). Inter-expert agreement was substantial to excellent for most Likert-scale criteria. These findings suggest that a hierarchical RAG architecture can provide reliable access to radiology-specific reference information, although external validation and automated updating of the knowledge base remain necessary. Full article

Accurate radiometric calibration is essential for high-resolution optical satellite sensors with limited onboard calibration capability, such as the FORMOSAT-5 (FS-5) Remote Sensing Instrument (RSI). The Radiometric Calibration Network (RadCalNet) provides standardized nadir-equivalent surface reflectance for vicarious calibration, but its direct application to off-nadir observations can introduce systematic biases over non-Lambertian surfaces. This study presents a BRDF-corrected vicarious calibration framework for the FS-5 RSI. The framework integrates RadCalNet data with an empirical BRDF lookup table built from in situ multi-angle measurements at Railroad Valley Playa, which is then propagated through 6S radiative transfer simulation. Applied to four FS-5 overpasses, BRDF correction reduced the median relative error of the calibration coefficient $K_0$ from 13–17% to 1–4% across all five spectral bands, providing a quantitative assessment of calibration improvement. The downstream impact was evaluated over an FS-5 La Crau scene. Scene-mean top-of-atmosphere (TOA) reflectance differences across the four multispectral bands ranged from 8.62% (NIR) to 10.99% (Green). The near-infrared reflectance of vegetation (NIRv), a proxy for gross primary production, showed a scene-mean relative difference of 7.88% ± 7.32%, with localized values exceeding 20% in densely vegetated areas. These results establish quantitative calibration-accuracy requirements for sensors relying on vicarious calibration and demonstrate the operational necessity of BRDF correction for reliable TOA reflectance and vegetation product retrieval. Full article

Tailings storage facility (TSF) failures have caused severe casualties and economic losses. This study used Enhanced Small Baseline Subset InSAR (E-SBAS-InSAR) and 88 Sentinel-1A images to retrieve the 2022–2024 surface deformation time series of the Shiguilong TSF, located in the Fe–Cu polymetallic metallogenic belt of the middle–lower Yangtze River. The reliability of the results was assessed through consistency comparisons with Small Baseline Subset InSAR (SBAS-InSAR) and Persistent Scatterer InSAR (PS-InSAR). A time-series decomposition model was applied to extract seasonal deformation components and analyze their lagged responses to temperature and intense rainfall events. The results show that: (1) E-SBAS-InSAR achieved a monitoring-point density nearly 7 times higher than SBAS-InSAR, enabling dense and long-term deformation characterization; (2) subsidence at Shiguilong continued to increase, with cumulative subsidence reaching −76.8 mm and a maximum annual mean subsidence rate of −22.78 mm/yr; (3) deformation was mainly controlled by long-term consolidation of loose tailings and creep of dam–tailings materials, while seasonal factors induced stage-dependent fluctuations; (4) seasonal deformation showed lagged responses of 6 days to temperature variations and 2 days to intense rainfall events, with rainfall exerting a more pronounced influence. This work is significant for TSFs monitoring under complex surface conditions. Full article

The trustworthiness of AI-powered network security auditing depends not only on detection accuracy but on the faithfulness of the explanations that support compliance verdicts. In IoT network security, Large Language Models (LLMs) are increasingly utilized to produce natural-language security assessments from raw network traffic, yet the extent to which these explanations are grounded in retrieved evidence is rarely measured. This paper presents the Retrieval-Augmented Generation Assessment Suite (RAGAS) as an evaluation framework that compares three retrieval paradigms—rule-based heuristic scoring, dense vector retrieval, and knowledge graph traversal—on the task of explaining network compliance against ETSI EN 303 645 IoT cybersecurity provisions. Using 30 human expert-validated compliance scenarios derived from the CIC-IoT2023 dataset and three LLMs (DeepSeek-R1, Qwen-2.5, Llama-3.2), we find that graph-based retrieval achieves the highest faithfulness (0.570), outperforming rule-based (0.524) and vector retrieval (0.509). All methods, however, exhibit low context recall (≤22.4%), and we highlight that high detection F1 scores do not guarantee faithful explanations; over 40% of statements in compliance answers are unsupported by retrieved evidence. A proof-of-concept prototype, Security Audit Compliance Agent (SACA), demonstrates how knowledge graph traversal can be integrated with interactive visualization to support human auditor oversight. We argue that, in adherence to responsible AI principles, faithfulness measurement should become a standard complement to accuracy reporting for an AI-driven network audit or forensic analysis. Full article

This paper evaluates a dense, total-variation-based optical flow method for retrieving wildfire smoke plume motion vectors from geostationary, deep-space, and airborne remote sensing imagery. Using multiple major fire events, we assess the robustness of the approach across a range of spatial resolutions and time intervals. The test cases include Geostationary Operational Environmental Satellite (GOES) observations of the 2025 Los Angeles Fires and the 2024 Park Fire, imagery from NASA’s Enhanced MODIS Airborne Simulator (eMAS) for the 2019 Sheridan and Williams Flats Fires, and a complementary Park Fire image pair from the Earth Polychromatic Imaging Camera (EPIC) aboard the Deep Space Climate Observatory (DSCOVR). Optical flow is computed directly on radiance fields, and smoke plumes are isolated using smoke masks derived from the Segmentation, Instance Tracking, and data Fusion Using multi-SEnsor imagery (SIT-FUSE) framework where available. Performance is evaluated by comparing the root mean square error (RMSE) between original image pairs and between the first image and the second image after warping with the retrieved motion field. RMSE is computed both globally and over smoke-only regions. Across GOES and eMAS cases, optical flow systematically reduces RMSE, often by more than a factor of two within smoke regions, indicating substantially improved frame-to-frame alignment of plume structures after motion correction. The DSCOVR/EPIC case, despite its coarser spatial resolution and longer temporal separation, also shows a marked reduction in global RMSE, demonstrating that the method remains informative under a broader range of observational conditions. For a selected subset of 10 consecutive GOES Park Fire pairs, we additionally compare the retrieved smoke motion vectors with collocated winds from the High-Resolution Rapid Refresh (HRRR) model and find the closest agreement in a broad lower-tropospheric layer centered near 875 hPa. These results show that dense optical flow can capture fine-scale plume evolution in high-temporal-resolution datasets while also providing useful motion estimates in coarser, global-view imagery. RMSE reduction is interpreted here as evidence of improved motion-compensated alignment, while the HRRR comparison provides initial physical context rather than independent validation. The resulting smoke motion vector fields provide a foundation for future comparison with model winds and for applications in plume analysis, fire hazard monitoring, and air quality studies. Full article

Rare-disease diagnosis is difficult because clinicians must identify plausible conditions from a large, severely imbalanced disease space using evidence distributed across clinical narratives, structured findings, and image-linked descriptions. This paper presents a hybrid pipeline with caption-mediated multimodal fusion for ranked rare-disease diagnosis and grounded explanation, developed and evaluated on the ZebraMap multimodal case-report dataset (69,146 structured cases; 1727 diseases). Grouped train–validation–test splitting by source article was applied to prevent leakage, and a sequential pipeline was constructed combining BM25 lexical retrieval, a class-balanced TF–IDF classifier, MedCPT dense retrieval and cross-encoder reranking, caption-based image-aware late fusion, and post hoc grounded explanation generation. The final pipeline achieved test MRR 0.3905 and Recall@10 0.5507 (nDCG@10 0.4273), while the strongest individual component, the class-balanced TF–IDF classifier, reached MRR 0.4200 and Recall@10 0.6279; the hybrid pipeline therefore integrates ranking with grounded explanation rather than maximizing single-metric diagnostic accuracy. On 256 explained cases, the explanation module achieved citation coverage 0.7334 and usefulness 3.8734, exposing a tradeoff between diagnostic accuracy and explanation richness. These results indicate that a hybrid retrieval-and-classification approach can support ranked rare-disease differential diagnosis and that grounded explanation quality can be evaluated quantitatively, extending computational support for the prolonged rare-disease diagnostic process. Full article

Complex agricultural question answering often requires multi-step reasoning over domain-specific knowledge and reliable retrieval of heterogeneous evidence. However, existing retrieval-augmented generation (RAG) methods usually rely on one-shot retrieval and provide limited control over whether the retrieved evidence is sufficient, accurate, and consistent for answering complex agricultural questions. To address this limitation, this paper proposes an agent-driven question-answering framework for the agricultural planting domain based on a Planning–Execution–Feedback (PEF) closed-loop mechanism. The framework decomposes complex queries into executable subtasks, performs knowledge acquisition through a knowledge-graph-guided hybrid retrieval module, and iteratively refines the reasoning process according to retrieval-quality feedback. Specifically, in the retrieval stage, a two-stage strategy is introduced to first localize candidate entities in the knowledge graph and then conduct context-enhanced dense retrieval with entity-consistency reranking, thereby reducing semantic drift and improving domain alignment. In the feedback stage, the agent evaluates the adequacy of the retrieved evidence and determines whether to continue execution, re-retrieve evidence, or replan the workflow. Experimental results on the AgroQA dataset show that the proposed method achieves 88.9%, 79.1%, and 92.6% on the Answer-C, Answer-R, and CR metrics, respectively, outperforming traditional retrieval-augmented and general large language model baselines. In addition, a three-dimensional digital human interface is implemented as an application prototype to demonstrate the feasibility of integrating the proposed framework into interactive agricultural knowledge services. Full article

Atmospheric correction is a critical preprocessing step in optical remote sensing, but repeated high-fidelity radiative transfer simulations remain computationally expensive for dense look-up-table generation, sensitivity analysis, retrieval support, and operational preprocessing. This study presents a physics-guided multi-fidelity surrogate framework for emulating atmospheric correction coefficients using paired 6S and libRadtran simulations. Atmospheric and geometric states are sampled using Latin Hypercube Sampling, and both radiative transfer models are evaluated under matched conditions for Sentinel-2 bands using spectral-response-function-aware coefficient generation. The high-fidelity targets are path reflectance, total transmittance, and spherical albedo. A physics-guided Kolmogorov–Arnold Network, termed pKANrtm, receives the atmospheric state and low-fidelity 6S coefficients, predicts the residual relative to libRadtran, and reconstructs the high-fidelity coefficients. The pKANrtm model uses an Efficient-KAN architecture and is trained with a physics-guided penalty applied in the original coefficient space. The proposed model is evaluated against state-of-the-art regression-based RTM surrogates. Across both standard and out-of-distribution (OOD) evaluation settings, pKANrtm achieves the strongest overall predictive performance among the compared models. Band-wise analysis shows that most Sentinel-2 bands are accurately emulated, while absorption-sensitive bands remain comparatively challenging. Runtime benchmarking demonstrates substantial acceleration relative to libRadtran, with GPU inference providing approximately four orders of magnitude single-sample speedup and batched inference reaching tens of thousands of samples per second. As an initial real-scene validation, the trained pKANrtm correction was applied to a Sentinel-2A acquisition over the Gobabeb RadCalNet site, demonstrating that the learned residual correction improves downstream surface-reflectance retrieval beyond synthetic RTM-to-RTM coefficient emulation. These results indicate that physics-guided multi-fidelity pKANrtm emulation provides an accurate, physically structured, computationally efficient, and practically useful strategy for atmospheric correction coefficient generation. Full article

This paper addresses the critical scalability challenge that large language model agents face when operating over massive tool repositories. As tool catalogs expand to hundreds or thousands of functions, current architectures exhibit substantial performance degradation caused by semantic collisions between similar tools and ineffective handling of complex multi-tool scenarios. To address these bottlenecks, we propose a recall-first Retrieval–Plan–Select (RPS) framework that combines context-aware query decomposition with synthetic tool description augmentation. The proposed approach explicitly separates retrieval, planning, and final selection through step-local candidate generation, while augmented tool descriptions enriched with expanded summaries and synthetic user questions reduce representation collisions in dense embedding spaces. Evaluation across Ultratool, ToolLinkOS, and ToolRet demonstrates that contextual decomposition consistently improves end-to-end recall under large tool catalogs, increasing recall from 0.340 to 0.494 on Ultratool, from 0.208 to 0.323 on ToolLinkOS, and from 0.300 to 0.347 on ToolRet. Description augmentation further improves retrieval quality, increasing Recall@10 from 0.288 to 0.403 and reducing high-similarity semantic collisions by 41.9% at the 0.90 cosine-similarity threshold. The proposed framework highlights that scalable tool use should be approached primarily as a recall-oriented retrieval and planning problem rather than as a flat in-context selection task, providing practical guidance for building large-scale tool-augmented agents over modern API and MCP-based ecosystems. Full article

Non-functional requirements (NFRs) are critical for ensuring software quality, yet they remain difficult to identify due to their implicit and loosely defined relationship with functional requirements (FRs). Existing research predominantly focuses on NFR classification, leaving the more practical problem of linking FRs with their corresponding NFRs largely underexplored. To bridge this gap, this research introduces Re-Distill, a framework that treats FR–NFR association as a retrieval task. It adopts a curriculum-guided, data-centric distillation strategy to improve semantic representations and capture the interdependencies between FRs and NFRs. The framework combines general semantic adaptation, domain-specific specialization, and teacher-guided hard-negative mining in a contrastive learning setting. During inference, it integrates dense and lexical retrieval with cross-encoder reranking to produce ranked NFR candidates for unseen FR queries. Experiments on an expanded FR–NFR dataset show consistent improvements throughout all training stages. The resulting model achieves a Recall@10 of 70.79%, significantly outperforming the zero-shot baseline (42.36% Recall@10). These results highlight the effectiveness of retrieval-based approaches for functional–non-functional requirement linking, providing a practical and scalable way to undertake software requirement analysis. Full article