Embedding-based generalized zero-shot learning (GZSL) models often first forge robust latent semantic correlations between visual and attribute features so that knowledge can generalize to unseen categories. Despite leveraging attributes as priors and learning a shared embedding space, current methods exhibit two critical flaws. First, attributes with heterogeneous granularity are treated uniformly, leading to semantic ambiguity. Second, the source of class-level misclassification seldom aligns with attribute-level errors, preventing models from targeting the specific attributes responsible. To overcome these limitations, we introduce Structured Attribute-Guided Enhancement (SAGE), a unified framework for GZSL. Consensus-aware bidirectional attention first synchronizes visual–semantic focus regions via a mutual-distillation scheme. Next, we partition all attributes into pairwise-disjoint subsets—Global, Context, and Local—and couple them with visual features extracted at matching spatial scales. Finally, we design a cross-sample, subset-aware distillation mechanism—when a sample is misclassified, SAGE identifies the culpable attribute subset, retrieves high-confidence prototypes from a memory bank, and applies a Kullback–Leibler (KL) divergence constraint to the corresponding feature branch. Comprehensive experiments and ablations on the challenging AwA2, CUB, and SUN benchmarks demonstrate the contribution of each component, with SAGE achieving a new state-of-the-art throughout. These findings underscore SAGE’s robustness and versatility, marking a substantial advance in generalized zero-shot learning and paving the way for broader zero-resource recognition.

Large Language Models (LLMs) have achieved remarkable success across a wide range of tasks, showcasing their transformative potential in natural language processing. However, these models face two persistent challenges: hallucination and forgetting. Hallucination occurs when models generate fabricated or inaccurate information in response to unfamiliar or out-of-distribution inputs, while forgetting refers to the inability to retain or recall previously learned knowledge during continued training. Although these phenomena have been extensively studied in isolation, their intrinsic connection remains largely un-explored. In this paper, we investigate the dialectical relationship between hallucination and forgetting through gradient dynamics during training. We propose a novel theory, Dynamic Cognitive Bias Entropy (HDCB), to jointly quantify the trade-offs between these two phenomena. To validate our findings, we conduct experiments based on two benchmark datasets, BIG-Bench Hard (BBH) and SQuAD, using LLaMA 3.2 1B and GPT 3.5 turbo models. Our results demonstrate that the order of training data significantly impacts the balance between hallucination and forgetting, with "known-first" training minimizing forgetting but increasing hallucination, and "unknown-first" training reducing hallucination but exacerbating forgetting. These findings, supported by entropy-based proofs and gradient-driven insights, provide a unified theoretical foundation and practical guidance for improving the reliability and robustness of LLMs.

Integrating language and images for Composed Image Retrieval (CIR) allows for a flexible representation of search intent, making it a focal point in multi-modal research. Traditional CIR methods train models on complex triplet datasets. In contrast, Zero-Shot Composed Image Retrieval (ZS-CIR) eliminates the need for constructing datasets and training models for each specific task, attracting significant attention from researchers. CIReVL is a training-free method that translates visual content into textual representations, employing large language models (LLMs) to perform robust text inference and visual language models (VLMs) to ensure multi-modal alignment during retrieval. Although this approach achieves high performance without training and offers strong interpretability, relying solely on single descriptive texts for retrieval has limitations in handling complex search demands. To address this issue, we propose a DualCIR framework, which utilizes an LLM to generate refined dual-directional descriptions-both positive and negative-alongside generating holistic target descriptions. We score retrieval data separately via these descriptions, merge the scores, perform ranking, and ultimately achieve retrieval. Extensive experiments across three datasets in the natural image and fashion domains show that our approach maintains low inference costs while noticeably enhancing retrieval effectiveness. Compared to existing training-free methods, our DualCIR achieves state-of-the-art performance and is even competitive with the training-based methods.

ABSTRACT In recent years, there has been an upward trend that marine vessels, an important object category in marine monitoring, have gradually become a research focal point in the field of computer vision, such as detection, tracking, and classification. Among them, marine vessel re‐identification (Re‐ID) emerges as a significant frontier research topics, which not only faces the dual challenge of huge intra‐class and small inter‐class differences, but also has complex environmental interference in the port monitoring scenarios. To propel advancements in marine vessel Re‐ID technology, SwinTransReID, a framework grounded in the Swin Transformer for marine vessel Re‐ID, is introduced. Specifically, the project initially encodes the triplet images separately as a sequence of blocks and construct a baseline model leveraging the Swin Transformer, achieving better performance on the Re‐ID benchmark dataset in comparison to convolution neural network (CNN)‐based approaches. And it introduces side information embedding (SIE) to further enhance the robust feature‐learning capabilities of Swin Transformer, thus, integrating non‐visual cues (orientation and type of vessel) and other auxiliary information (hull colour) through the insertion of learnable embedding modules. Additionally, the project presents VesselReID‐1656, the first annotated large‐scale benchmark dataset for vessel Re‐ID in real‐world ocean surveillance, comprising 135,866 images of 1656 vessels along with 5 orientations, 12 types, and 17 colours. The proposed method achieves 87.1 mAP and 96.1 Rank‐1 accuracy on the newly‐labelled challenging dataset, which surpasses the state‐of‐the‐art (SOTA) method by 1.9 mAP regarding to performance. Moreover, extensive empirical results demonstrate the superiority of the proposed SwinTransReID on the person Market‐1501 dataset, vehicle VeRi‐776 dataset, and Boat Re‐ID vessel dataset.

This article proposes a shared control system model between the human pilot and autopilot for the special issue of carrier-based aircraft landing task. A key point of the shared and cooperative control is that the decision-sharing system depends on the longitudinal safety boundaries for manual/automatic landing. Two strategies of the human pilot are adopted, including capture strategy and tracking strategy. A hidden model tracking control method is utilized to model the autopilot. To address the issue of frequent switching between the human pilot and autopilot caused by relying solely on safety boundaries to allocate control authority, fuzzy control theory is introduced to reduce the workload of the human pilot. The time-domain simulation results show that considering the fuzzy control, the frequency of switching and the flight states have been improved compared with the results without fuzzy control. Nonlinear pilot-induced oscillation metric evaluation results show that the human-automation shared and cooperative control considering the fuzzy control can alleviate the workload of the human pilot. The shared and cooperative control system model has certain significance in ensuring the safety of carrier-based aircraft landing.

The shipping industry has experienced rapid growth in recent years, prompting a need for advanced target recognition technology based on marine radar. This paper introduces the Track Classification Model (TCM), a novel approach for classifying track sequences in real scenarios. The TCM utilizes a feature extraction network based on multi-feature fusion, taking radar echo images and motion information of the target as input, to improve classification accuracy. Additionally, the paper also presents a dataset production method that addresses the issue of missing labels, a critical problem in track sequence classification. Through ablation experiments, the paper demonstrates the effectiveness of the design strategy, with the multi-feature fusion network successfully extracting features and achieving superior performance over single feature extraction networks. The results show that increasing the number of input track points and raising the upper limit of the input sequence leads to improved classification accuracy. Finally, in real scenarios, the proposed model outperforms other algorithms, showcasing its high engineering application value.

This paper presents a digitally programmable bi-directional 7-bit passive phase shifter in a 65 nm CMOS technology. The core of this passive vector-synthesized phase shifter is a hybrid quadrature generator (HQG), an interstage matching network, and a passive vector modulator (PVM). This work proposes a high coupling-factor-based quadrature generator design methodology and demonstrates it with a compact vertical transformer. The interstage matching network between HQG and PVM is proposed to release the bandwidth bottleneck and achieve a 34% fractional frequency bandwidth. Two 6-bit X-type attenuators in the I and Q path form a high-resolution 12-bit controlling word. In 32–40 GHz, this 7-bit 360° phase shifter achieves a measured 2.8° step with 0.45-1.6° RMS phase error and 0.2-0.36 dB RMS magnitude error. With the broadband technique, its 3-dB bandwidth reaches 30.2-42.7 GHz with a 2.8° RMS phase error. Its in-band 1-dB compression point is 10.2 dBm. With the proposed compact HQG and PVM, this mm-wave passive phase shifter only occupies$220\times 630\,\,\mu \text{m}^{2}$and has no power consumption.

This paper presents a fully-integrated 76-81 GHz four-channel digitally controlled receiver (RX) in 65-nm CMOS, which utilizes a four-channel RX front-end with high linearity, an LO distribution network, and a reconfigurable 4-channel analog baseband (ABB). It achieves high integration level and is capable of flexible reconfigurability automotive radars with wide and flattened temperature characteristic. Each RX front-end consists of a low noise amplifier (LNA) employing a neutralized common source (CS) with configurable high-linearity and an active mixer with linearity-priority design. Additionally, a bias circuit with positive temperature coefficient is used for temperature compensation of the millimeter wave (mm-Wave) amplifiers and mixer. The RX chain offers tunable gain and bandwidth (BW) with digital control and bandwidth tuning for the ABB. The measured results show that the RX achieves an input 1-dB compression point (Pin,1dB) of -7 dBm and a noise figure of 11-26 dB across a temperature range from -45 to +125°C, both over the range of 76-81 GHz. The RX has a tunable gain from 18 to 66 dB over the 3-dB programmable IF BW from 0.1 to 10 MHz. The power consumption of the entire receiver is 0.52 W.

This paper presents a 77 GHz Receiver (Rx) for automotive radar with high linearity, low noise figure (NF) and reconfigurable gain and bandwidth of the analog baseband (ABB), implemented in TSMC 65 nm CMOS general purpose (GP) technology. The Rx includes a low-noise amplifier (LNA), sharing the transconductance stage with a double-balanced mixer, an LO chain with a frequency doubler and an analog baseband. The entire Rx consumes 123 mW. Measurement results show that the input 1 dB compression point (P1dB) is -14 dBm at 81 GHz, a conversion gain (CG) of 20.5-62.5 dB is digitally tunable at 77 GHz, and an NF of 11.2-18.3 dB is achieved over the frequency range of 76-81 GHz.