Yi Li

Yi Li

Auto-generated profile, based on publicly available information

Yi Li

College of Electronic Engineering, Naval University of Engineering, Wuhan, Hubei, China

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 June 2023 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
HLHailiang Lu
Hailiang Lu
YLYinan Li
Yinan Li
XXXu Xie
Xu Xie
YLYi Li
Yi Li
QFQingbiao Fan
Qingbiao Fan
QLQingxia Li
Qingxia Li
The critical advantage of passive microwave imaging is that it can be used in all-day and in low visibility conditions, such as hazes, fogs, clouds, smokes, and sandstorms. In this article, the ground-based passive interferometric microwave sensor (PIMS) is proposed to detect civil aircrafts, especially in low visibility conditions. The fundamental of civil aircraft detection by a ground-based PIMS is introduced and analyzed. Numerical simulations are performed to assess its feasibility and detection performance, and the impact of the low visibility weather is analyzed and discussed. To verify the practicability of civil aircraft detection by the ground-based PIMS, a series of experiments have been performed by a ground-based X-band PIMS, and a data processing procedure is introduced, which contains a target detection algorithm. Experimental results have demonstrated that the civil aircrafts can be accurately detected and consecutively tracked by the ground-based PIMS using the target detection algorithm. Meanwhile, a mirrored TB phenomenon, the TB change, and an application scheme are analyzed and discussed. As a conclusion, the ground-based PIMS can be considered as a powerful complementary tool for civil aircraft detection.

20 March 2026
Verified
Conference Paper
Open access
HZHuatao Zhou
Huatao Zhou
LZLi Zhao
Li Zhao
LTLingli Tan
Lingli Tan
YLYi Li
Yi Li
With the rapid development of artificial intelligence technology, large language models face significant challenges when processing professional domain knowledge, including outdated knowledge and hallucination phenomena. This paper presents a comprehensive evaluation of a professional domain RAG knowledge Q&A system built using DeepSeek-R1-7B and mxbai-embed-large on Huawei Developer Space cloud infrastructure. We conduct detailed ablation studies to quantify the contribution of each system component, propose an efficient incremental knowledge update workflow for dynamic knowledge bases, and systematically measure hallucination rates using automated fact-checking against ground truth annotations. Benchmark comparisons against three competitive baselines including LlamaIndex, LangChain native RAG, and vanilla LLM demonstrate that our system achieves 23.5% higher retrieval recall@5 and 31.2% improvement in end-to-end QA accuracy. Experimental results validate that RAG technology effectively mitigates hallucination by 67.4% compared to baseline LLM without retrieval augmentation. This study provides a feasible technical solution with rigorous quantitative evaluation methodology for organizations seeking to build private knowledge Q&A systems.