Improving Health Literacy through Lay Summarization of Radiological Reports: An Evaluation of BioNER and Retrieval-Augmented Generation
研究通过结合生物命名实体识别（NER）和检索增强生成（RAG）的方法，提高放射学报告对患者的可读性和准确性，解决患者难以理解专业术语的问题。
0 citationsRead paper
Selected work
Representative Papers
RLG-TPV: Radar- and LiDAR-Guided Tri-Perspective View Fusion for Camera-Radar 3D Object Detection
本文提出RLG-TPV框架，利用雷达和激光雷达提供几何指导，解决相机-雷达3D目标检测中的深度模糊问题。
0 citationsRead paper
RetrievalRouter: Joint Modality and Architecture Selection for Document Retrieval
本文提出RetrievalRouter，通过分析查询文本选择最优检索管道，解决文档检索中准确度与速度之间的矛盾。
0 citationsRead paper
Evaluating Modern RAG: Textual, Multimodal, Dense, and Late Interaction Pipelines
本文提出一种数据驱动的选择方法，帮助选择适合文档语料库的RAG管道，通过评估文本和多模态管道来平衡检索性能与系统效率。
0 citationsRead paper
OVS Meets PQ-TLS: Exploring Post-Quantum TLS for SDN's Southbound API
This study addresses the quantum computing threats facing SDN southbound APIs by developing a prototype system supporting post-quantum TLS and conducting the first systematic performance evaluation in both pure and hybrid modes. By quantifying the latency and CPU overhead of various post-quantum signature and key exchange schemes, this work identifies optimal algorithm combinations for deployment. The research fills a critical empirical gap regarding PQ-TLS implementation within SDN southbound interfaces, establishing essential performance benchmarks and providing a technical foundation for the quantum-resistant migration of software-defined networks. These findings offer actionable guidance for securing next-generation network infrastructures against emerging cryptographic vulnerabilities while maintaining operational efficiency in control plane communications.
0 citationsRead paper
Recent publications
Latest Papers
Improving Health Literacy through Lay Summarization of Radiological Reports: An Evaluation of BioNER and Retrieval-Augmented Generation
研究通过结合生物命名实体识别（NER）和检索增强生成（RAG）的方法，提高放射学报告对患者的可读性和准确性，解决患者难以理解专业术语的问题。
0 citationsRead paper
RLG-TPV: Radar- and LiDAR-Guided Tri-Perspective View Fusion for Camera-Radar 3D Object Detection
本文提出RLG-TPV框架，利用雷达和激光雷达提供几何指导，解决相机-雷达3D目标检测中的深度模糊问题。
0 citationsRead paper
RetrievalRouter: Joint Modality and Architecture Selection for Document Retrieval
本文提出RetrievalRouter，通过分析查询文本选择最优检索管道，解决文档检索中准确度与速度之间的矛盾。
0 citationsRead paper
Evaluating Modern RAG: Textual, Multimodal, Dense, and Late Interaction Pipelines
本文提出一种数据驱动的选择方法，帮助选择适合文档语料库的RAG管道，通过评估文本和多模态管道来平衡检索性能与系统效率。
0 citationsRead paper
OVS Meets PQ-TLS: Exploring Post-Quantum TLS for SDN's Southbound API
This study addresses the quantum computing threats facing SDN southbound APIs by developing a prototype system supporting post-quantum TLS and conducting the first systematic performance evaluation in both pure and hybrid modes. By quantifying the latency and CPU overhead of various post-quantum signature and key exchange schemes, this work identifies optimal algorithm combinations for deployment. The research fills a critical empirical gap regarding PQ-TLS implementation within SDN southbound interfaces, establishing essential performance benchmarks and providing a technical foundation for the quantum-resistant migration of software-defined networks. These findings offer actionable guidance for securing next-generation network infrastructures against emerging cryptographic vulnerabilities while maintaining operational efficiency in control plane communications.
0 citationsRead paper