Malicious URL detection faces persistent challenges from morphological heterogeneity and frequent concept drift. Traditional deep learning models rely on static decision boundaries and require computationally expensive retraining to counter emerging threats, creating a dangerous detection lag. To bridge this research gap, we propose MLF-ICL, a novel detection framework that shifts the paradigm from static classification to an inference model that utilizes retrieval and requires no training. To enable this transition and connect raw text with tabular models, we introduce a specialized architecture designed to bridge features. First, we engineer a DCG-BERT backbone enhanced by DFF to capture microscopic character obfuscations and macroscopic token semantics. Second, we develop the M-DWFS algorithm to dynamically fuse these deep representations with lexical rules derived from expert knowledge. This highly accurate multimodal feature space allows us to pioneer the application of a Tabular Foundation Model (TabDPT) equipped with In-Context Learning (ICL). By replacing parameter updates with reasoning based on retrieval, our model dynamically adapts to new threat variants using historical context. Extensive evaluations on three benchmark datasets demonstrate that MLF-ICL achieves superior performance, including an F1 score of 99.78% on highly imbalanced data. The results validate the proposed paradigm as a highly effective solution for evolving cybersecurity challenges.

Modern engineering, spanning electrical, mechanical, aerospace, civil, and computer disciplines, stands as a cornerstone of human civilization and the foundation of our society. However, engineering design poses a fundamentally different challenge for large language models (LLMs) compared with traditional textbook-style problem solving or factual question answering. Although existing benchmarks have driven progress in areas such as language understanding, code synthesis, and scientific problem solving, real-world engineering design demands the synthesis of domain knowledge, navigation of complex trade-offs, and management of the tedious processes that consume much of practicing engineers' time. Despite these shared challenges across engineering disciplines, no benchmark currently captures the unique demands of engineering design work. In this work, we introduce EngDesign, an Engineering Design benchmark that evaluates LLMs' abilities to perform practical design tasks across nine engineering domains. Unlike existing benchmarks that focus on factual recall or question answering, EngDesign uniquely emphasizes LLMs' ability to synthesize domain knowledge, reason under constraints, and generate functional, objective-oriented engineering designs. Each task in EngDesign represents a real-world engineering design problem, accompanied by a detailed task description specifying design goals, constraints, and performance requirements. EngDesign pioneers a simulation-based evaluation paradigm that moves beyond textbook knowledge to assess genuine engineering design capabilities and shifts evaluation from static answer checking to dynamic, simulation-driven functional verification, marking a crucial step toward realizing the vision of engineering Artificial General Intelligence (AGI).

With the development of multimodality and large language models, the deep learning-based technique for medical image captioning holds the potential to offer valuable diagnostic recommendations. However, current generic text and image pre-trained models do not yield satisfactory results when it comes to describing intricate details within medical images. In this paper, we present a novel medical image captioning method guided by the segment anything model (SAM) to enable enhanced encoding with both general and detailed feature extraction. In addition, our approach employs a distinctive pre-training strategy with mixed semantic learning to simultaneously capture both the overall information and finer details within medical images. We demonstrate the effectiveness of this approach, as it outperforms the pre-trained BLIP2 model on various evaluation metrics for generating descriptions of medical images.