Haitao Wang

Large Language Models (LLMs) have demonstrated extraordinary capabilities in natural language generation; however, their real-world deployment is frequently hindered by the generation of factually incorrect or biased content, along with an inherent deficiency in transparency. To address these critical limitations and thereby enhance the reliability and explainability of LLM outputs, this study proposes a novel integrated framework, namely the Adaptive Knowledge-Driven Correction Network (AKDC-Net), which incorporates three core algorithmic innovations. Firstly, the Hierarchical Uncertainty-Aware Bias Detector (HUABD) performs multi-level linguistic analysis (lexical, syntactic, semantic, and pragmatic) and, for the first time, decomposes predictive uncertainty into epistemic and aleatoric components. This decomposition enables principled, interpretable bias detection with clear theoretical underpinnings. Secondly, the Neural-Symbolic Knowledge Graph Enhanced Corrector (NSKGEC) integrates a temporal graph neural network with a differentiable symbolic reasoning module, facilitating logically consistent and factually grounded corrections based on dynamically updated knowledge sources. Thirdly, the Contrastive Learning-driven Multimodal Explanation Generator (CLMEG) leverages a cross-modal attention mechanism within a contrastive learning paradigm to generate coherent, high-quality textual and visual explanations that enhance the interpretability of LLM outputs. Extensive evaluations were conducted on a challenging medical domain dataset to validate the effectiveness of the proposed AKDC-Net framework. Experimental results demonstrate significant improvements over state-of-the-art baselines: specifically, a 14.1% increase in the F1-score for bias detection, a 19.4% enhancement in correction quality, and a 31.4% rise in user trust scores. These findings establish a new benchmark for the development of more trustworthy and transparent artificial intelligence (AI) systems, laying a solid foundation for the broader and more reliable application of LLMs in high-stakes domains.

Domain shifts pose significant challenges for cross-domain semantic segmentation in high-resolution remote sensing imagery. Inspired by the cognitive mechanisms of the human brain, we propose a Brain-Inspired Style Transfer and Semantic Segmentation Collaborative Adversarial Framework (SAF), which mimics neural processes such as hierarchical perception, memory retrieval, and multimodal integration to enhance cross-domain feature alignment and segmentation performance. To achieve the joint optimization of style transfer and semantic segmentation networks, we introduce three key components: a Semantic-Aware Transformer Module (SATM) that dynamically captures and preserves essential semantic features during style transfer; a Semantic-Driven Multi-feature Memory Module (SMM) that stores and retrieves historical style and semantic information to improve domain adaptability; a Domain-Invariant Style-Semantic Center Space (DSCS) that aligns style and semantic features within a shared representation space, mitigating discrepancies between style and semantic domains. Extensive experiments across multiple tasks demonstrate that SAF effectively reduces distortions and semantic inconsistencies by achieving deep style–semantic alignment. Compared to leading approaches, SAF achieves a superior balance between style adaptation and semantic preservation, significantly improving model generalization in remote sensing applications.

The reactor pressure vessel (RPV) is a critical barrier in nuclear power plants, but its embrittlement during service poses a significant safety challenge. This study investigated the effects of Cu-enriched clusters on the mechanical and magnetic properties of Fe-0.9 wt.%Cu model alloys through thermal aging. Using Vickers hardness tests, Magnetic Barkhausen Noise (MBN) detection, and Atom Probe Tomography (APT), the study aimed to establish a quantitative correlation between MBN signals, Vickers hardness, and Cu-enriched clusters, facilitating the non-destructive testing of RPV embrittlement. Experimental results showed that the hardness and MBN parameters (RMS and V_pp values) changed significantly with aging time. The hardness increased rapidly in the early stage (under-aged), followed by a plateau and then a decreasing trend (over-aged). In contrast, MBN parameters decreased initially and then increased. APT analysis revealed that Cu-enriched clusters increase in size to 4.60 nm and coalesced during aging, with their number density peaking to 3.76 × 10²³ m⁻³ before declining. An inverse linear correlation was found between MBN signals and the combined factor N_d²R_g (product of the number density squared and the mean radius of Cu-enriched clusters). This correlation was consistent across both under-aged and over-aged states, suggesting that MBN signals can serve as applicable indicators for the non-destructive evaluation of RPV steel embrittlement.

Pixel-level classification of very-high-resolution images is a crucial yet challenging task in remote sensing. While transformers have demonstrated effectiveness in capturing dependencies, their tendency to partition images into patches may restrict their applicability to highly detailed remote sensing images. To extract latent contextual semantic information from high-resolution remote sensing images, we proposed a gaze–saccade transformer (GSV-Trans) with visual perceptual attention. GSV-Trans incorporates a visual perceptual attention (VPA) mechanism that dynamically allocates computational resources based on the semantic complexity of the image. The VPA mechanism includes both gaze attention and eye movement attention, enabling the model to focus on the most critical parts of the image and acquire competitive semantic information. Additionally, to capture contextual semantic information across different levels in the image, we designed an inter-layer short-term visual memory module with bidirectional affinity propagation to guide attention allocation. Furthermore, we introduced a dual-branch pseudo-label module (DBPL) that imposes pixel-level and category-level semantic constraints on both gaze and saccade branches. DBPL encourages the model to extract domain-invariant features and align semantic information across different domains in the feature space. Extensive experiments on multiple pixel-level classification benchmarks confirm the effectiveness and superiority of our method over the state of the art.

Synthetic aperture radar (SAR) image ship detection is currently a research hotspot in the field of national defense science and technology. However, SAR images contain a large amount of coherent speckle noise, which poses significant challenges in the task of ship detection. To address this issue, we propose filter convolution, a novel design that replaces the traditional convolution layer and suppresses coherent speckle noise while extracting features. Specifically, the convolution kernel of the filter convolution comes from the input and is generated by two modules: the kernel-generation module and local weight generation module. The kernel-generation module is a dynamic structure that generates dynamic convolution kernels using input image or feature information. The local weight generation module is based on the statistical characteristics of the input images or features and is used to generate local weights. The introduction of local weights allows the extracted features to contain more local characteristic information, which is conducive to ship detection in SAR images. In addition, we proved that the fusion of the proposed kernel-generation module and the local weight module can suppress coherent speckle noise in the SAR image. The experimental results show the excellent performance of our method on a large-scale SAR ship detection dataset-v1.0 (LS-SSDD-v1.0). It also achieved state-of-the-art performance on a high-resolution SAR image dataset (HRSID), which confirmed its applicability.

Pulsed Eddy Current (PEC) NDT has played a vital role in detection and classification of the surface and sub-surface defects in conductive structures. Normally, it uses peak values of the acquired transient field signals, and the combination of the feature values of the time of the peak to identify flaws with the help of Principal Component Analysis (PCA). However, it is found that the random noise undermines the classification results, because PCA works robustly only in the time domain. In the light of this drawback, the fundamental and the first-harmonic components are investigated and taken as the new feature values in the frequency domain. Through the analysis of the feature values in both time and frequency domains, the influence of random noise is mitigated. Consequently, surface defects, subsurface defects and metal thickness changes are classified with much higher identification accuracy.

Remote-field eddy current (RFEC) inspection has shown its advantages over the other techniques in Electromagnetic non-destructive evaluation (ENDE) techniques in detection and characterization of deeper defects. In this paper, a new RFEC transducer has been designed for inspection of flat conductive plates, based on numerical simulations using finite element method. Through experimental study, it is concluded that the new transducer has the capability of detecting deep buried defects, and that the phase of RFEC signals have linear relation with the depth of the defects.