Large language models (LLMs) are increasingly applied in specialized domains such as finance and healthcare, where they introduce unique safety risks.Domain-specific datasets of harmful prompts remain scarce and still largely rely on manual construction; public datasets mainly focus on explicit harmful prompts, which modern LLM defenses can often detect and refuse.In contrast, implicit harmful prompts-expressed through indirect domain knowledge-are harder to detect and better reflect real-world threats.We identify two challenges: transforming domain knowledge into actionable constraints and increasing the implicitness of generated harmful prompts.To address them, we propose an end-to-end framework that first performs knowledge-graphguided harmful prompt generation to systematically produce domain-relevant prompts, and then applies two-strategy obfuscation rewriting to convert explicit harmful prompts into implicit variants via direct and context-enhanced rewriting.This framework yields high-quality datasets combining strong domain relevance with implicitness, enabling more realistic redteaming and advancing LLM safety research.We release our code and datasets on GitHub.

Computational notebooks offer a flexible environment for exploratory data analysis (EDA), but this flexibility often leads to disorganized and iterative execution of notebook cells, making it difficult to track how data states evolve. Consequently, data scientists must devote extra mental effort to staying aware of data states, which is both tedious and prone to overlooking anomalies. To address this challenge, we developed NoteFlow, a notebook extension that leverages charts as “sight glasses” to provide a consistent and continuous tracing of data flow. NoteFlow allows users to (1) validate various facets of the current data state using recommended charts provided immediately after each cell execution, and (2) trace the global evolution of selected charts to continuously observe how particular data attributes evolve throughout the EDA process. We evaluated NoteFlow’s effectiveness through a controlled study with 12 participants and a one-month field study with 2 data scientists on real-world workflows.

ABSTRACT The detection of small objects in remote sensing imagery presents significant challenges, mainly attributed to intricate background complexities, inadequate feature extraction performance, and constrained receptive field dimensions. Moreover, under constrained computational resources, improving detection accuracy while reducing the model's computational complexity and parameter count has become an urgent issue. In response to these challenges, a novel lightweight model specifically designed for small object detection, named DWT‐YOLO, is proposed in this study. By introducing the Discrete Wavelet Transform (DWT) combined with convolutional down‐sampling operations, the model effectively expands the receptive field and enhances feature extraction capabilities. Simultaneously, an innovative Wavelet Convolution Fusion module (WCFM) is designed, which improves multi‐scale feature fusion and cross‐domain information association, enhancing the feature representation for small objects while keeping model complexity relatively low. Additionally, the model incorporates a Spatial‐Channel Decoupling Down‐sampling (SCDown) module, which effectively reduces redundant parameters and improves detection accuracy. We first validated the effectiveness of DWT‐YOLO on the RSOD dataset and then further assessed its small object detection performance on the VisDrone2019 dataset. The proposed model demonstrates a 5.4% enhancement in detection accuracy compared to the baseline model. The model has only 5.8M parameters, and it outperforms several advanced models in detection performance. Experimental results demonstrate that DWT‐YOLO not only achieves significant improvements in small object detection tasks but also exhibits lightweight characteristics, providing new solutions and technical insights for this field. The implementation source code of the proposed method is publicly accessible at the following repository: http://github.com/zzgithubFly/DWT‐YOLO .

Large Language Models (LLMs) are powerful but also raise significant security concerns, particularly regarding the harm they can cause, such as generating fake news that manipulates public opinion on social media and providing responses to unethical activities. Traditional red teaming approaches for identifying AI vulnerabilities rely on manual prompt construction and expertise. This paper introduces AdversaFlow, a novel visual analytics system designed to enhance LLM security against adversarial attacks through human-AI collaboration. AdversaFlow involves adversarial training between a target model and a red model, featuring unique multi-level adversarial flow and fluctuation path visualizations. These features provide insights into adversarial dynamics and LLM robustness, enabling experts to identify and mitigate vulnerabilities effectively. We present quantitative evaluations and case studies validating our system's utility and offering insights for future AI security solutions. Our method can enhance LLM security, supporting downstream scenarios like social media regulation by enabling more effective detection, monitoring, and mitigation of harmful content and behaviors.