Generative multi-domain transfer learning for fault detection in data-scarce wind turbines
本文针对数据稀缺风力发电机故障检测问题，提出基于StarGAN的多域生成映射方法，以改善故障检测性能。
0 citationsRead paper
Selected work
Representative Papers
Breaking Darknet CAPTCHAs with general purpose LLM
研究评估了通用大语言模型破解暗网验证码的有效性，通过结合经典计算机视觉工具克服其在精确定位和几何变换上的不足。
0 citationsRead paper
Tropospheric temperature and humidity profile retrieval from Meteosat Flexible Combined Imager based on deep learning
研究利用深度学习方法，特别是空间感知的Residual U-Net模型，从Meteosat Flexible Combined Imager数据中独立提取对流层温度和湿度剖面，不依赖数值天气预报背景场。
0 citationsRead paper
Who Owns the Online Media?
This study addresses the challenges of media ownership opacity and accountability by constructing ownership networks for thousands of European and American outlets to measure transnational market concentration. Employing a fixed-pair design, we examine the causal effect of ownership changes on content similarity. Results indicate that over half of media entities are controlled by single owners, with common ownership significantly driving reporting convergence, particularly in the United States. The primary contribution lies in establishing cross-nationally comparable quantitative metrics demonstrating that content homogenization is predominantly driven by ownership structures rather than audience demand. These findings provide robust causal evidence regarding the detrimental impact of media consolidation on information diversity, clarifying the structural determinants of news uniformity across Western media markets.
0 citationsRead paper
Reinforced Graph of Thoughts: RL-Driven Adaptive Prompting for LLMs
This work addresses the limitation of traditional Graph of Thoughts frameworks, which rely on manually predefined operation graphs and struggle to adapt to varying task complexity. To overcome this, the paper introduces a novel approach that integrates large language models with reinforcement learning to automatically construct task-adaptive reasoning operation graphs from a human-defined set of operations. By leveraging reinforcement learning, the method dynamically generates operation graphs tailored to the specific demands of each task, moving beyond the constraints of static, handcrafted designs. This enables adaptive graph construction under given constraints, significantly enhancing both the flexibility and performance of complex problem-solving. The proposed framework represents the first application of reinforcement learning within the Graph of Thoughts paradigm, offering a principled and scalable solution for automating reasoning structures.
0 citationsRead paper
Recent publications
Latest Papers
Generative multi-domain transfer learning for fault detection in data-scarce wind turbines
本文针对数据稀缺风力发电机故障检测问题，提出基于StarGAN的多域生成映射方法，以改善故障检测性能。
0 citationsRead paper
Breaking Darknet CAPTCHAs with general purpose LLM
研究评估了通用大语言模型破解暗网验证码的有效性，通过结合经典计算机视觉工具克服其在精确定位和几何变换上的不足。
0 citationsRead paper
Tropospheric temperature and humidity profile retrieval from Meteosat Flexible Combined Imager based on deep learning
研究利用深度学习方法，特别是空间感知的Residual U-Net模型，从Meteosat Flexible Combined Imager数据中独立提取对流层温度和湿度剖面，不依赖数值天气预报背景场。
0 citationsRead paper
Who Owns the Online Media?
This study addresses the challenges of media ownership opacity and accountability by constructing ownership networks for thousands of European and American outlets to measure transnational market concentration. Employing a fixed-pair design, we examine the causal effect of ownership changes on content similarity. Results indicate that over half of media entities are controlled by single owners, with common ownership significantly driving reporting convergence, particularly in the United States. The primary contribution lies in establishing cross-nationally comparable quantitative metrics demonstrating that content homogenization is predominantly driven by ownership structures rather than audience demand. These findings provide robust causal evidence regarding the detrimental impact of media consolidation on information diversity, clarifying the structural determinants of news uniformity across Western media markets.
0 citationsRead paper
Reinforced Graph of Thoughts: RL-Driven Adaptive Prompting for LLMs
This work addresses the limitation of traditional Graph of Thoughts frameworks, which rely on manually predefined operation graphs and struggle to adapt to varying task complexity. To overcome this, the paper introduces a novel approach that integrates large language models with reinforcement learning to automatically construct task-adaptive reasoning operation graphs from a human-defined set of operations. By leveraging reinforcement learning, the method dynamically generates operation graphs tailored to the specific demands of each task, moving beyond the constraints of static, handcrafted designs. This enables adaptive graph construction under given constraints, significantly enhancing both the flexibility and performance of complex problem-solving. The proposed framework represents the first application of reinforcement learning within the Graph of Thoughts paradigm, offering a principled and scalable solution for automating reasoning structures.
0 citationsRead paper