Shi Cheng

Mr. Shi Cheng

orcid

Research Focus & Metrics

Academic Career Length

19Years (2008-Present)

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

Representative Publications

View All Publications
20 May 2025 in Journal of King Saud University Computer and Information Sciences
Verified
Journal Article
Open access
PDF file
XCXiaoyong Chen
Xiaoyong Chen
ZCZhiqiang Chen
Zhiqiang Chen
SCShi Cheng
Shi Cheng
Long-document summarization remains a challenging task for large language models (LLMs), which often suffer from input length constraints, semantic incoherence, and factual hallucinations when processing extensive and complex texts. In this paper, we propose a novel summarization framework that integrates hierarchical input segmentation with Chain-of-Thought (CoT) prompting to guide LLMs through structured, interpretable reasoning. Our method decomposes long documents into semantically coherent segments, applies CoT-based prompting for intermediate summary reasoning, and employs structure-guided decoding to compose high-quality final summaries. We evaluate our approach across five diverse datasets, including scientific, biomedical, governmental, literary, and legal domains, using strong LLM backbones such as Qwen, LLaMA, and Phi. Experimental results demonstrate that our method consistently outperforms state-of-the-art baselines across ROUGE, BLEU, BERTScore, and factual consistency metrics. Ablation and human evaluation further confirm the complementary benefits of CoT reasoning and hierarchical structure, offering a reliable and scalable solution for summarizing complex long-form content.

12 May 2023 in Signal, Image and Video Processing
Verified
Journal Article
Closed access
PDF file
HBHu Bin
Hu Bin
LJLi Jinhang
Li Jinhang
ZLZhao Lili
Zhao Lili
SCShi Cheng
Shi Cheng

18 July 2023 in Applied Sciences
Verified
Journal Article
Open access
PDF file
XWXusen Wan
Xusen Wan
JZJinbao Zhang
Jinbao Zhang
SCShi Cheng
Shi Cheng
WWWeixiang Wu
Weixiang Wu
JWJiehua Wang
Jiehua Wang
Differential fault attack (DFA) is a distinctive methodology for acquiring the key to block ciphers, which comprises two distinct strategies: DFA on the state and DFA on the key schedule. Given the widespread adoption of the Advanced Encryption Standard (AES), it has emerged as a prominent target for DFA. This paper presents an efficient DFA on the AES, utilizing a two−byte fault model that induces faults at the state with discontiguous rows. The experiment demonstrates that, based on the proposed fault model, the key for AES–128, AES–192, and AES–256 can be successfully recovered by exploiting two, two, and four faults, respectively, without the need for exhaustive research. Notably, in the case of AES–256, when considering exhaustive research, two (or three) faults are needed with 232 (or 216) exhaustive searches. In comparison to the currently available DFA on the AES state, the proposed attack method shows a higher efficiency due to the reduced induced faults.

1 January 2024 in Trans/Form/Ação
Verified
Journal Article
Open access
PDF file
ZXZichen Xu
Zichen Xu
SCShi Cheng
Shi Cheng
With the modern information society’s rapid development, Big Data (BD) has brought great convenience to residents’ lives. Under the background of BD, the application of information technology in various fields has become a trend. Through the transformation of educational philosophy, Higher Education Management (HEM) has also taken on new characteristics and generated new development directions. This article explored the shortcomings and reasons of traditional HEM work, pointed out current problems in it and explained the necessity of innovation in it. This article also analyzed the advantages and development of BD technology in management work. Through the new ideas, methods and tools brought by Big Data technology to Higher Education Management, it was demonstrated that BD technology supported the development of innovative teaching mode, provided a positive learning environment and reduces management costs. Based on the characteristics of HEM work and the advantages of BD technology, innovative research was conducted to analyze students’ learning efficiency and enthusiasm to illustrate the innovative effect of HEM work. BD technology, including Artificial Intelligence and precise and flexible management, brings new ideas, methods and tools to HEM, effectively improving efficiency, enhancing students’ happiness and reducing labor management costs. They support the development of innovative teaching models, offer university students a positive learning environment and cut down on management costs.

1 January 2008
Verified
Conference Paper
Closed access
GWGuoping Wu
Guoping Wu
SCShi Cheng
Shi Cheng
MAMinsi Ao
Minsi Ao
LHLei Hui
Lei Hui
Slant Correction plays an important role in vehicle license plate automatic recognition system. In order to reduce or avoid adverse effects caused by noise disturbance and fragmentary frame of vehicle license plate, as well as speed up the computation, an approach based on feature point and principal component analysis (PCA) is presented. Feature points are considered as the edge of characters on the license plate, with a lined orderliness which reflects the slant angle of the plate. Firstly, a pretreatment process is carried out to extract the feature points of the plate, and the direction of the principal component, which is considered as the slant angle of the plate, is achieved through principal component analysis of the feature points, then the correction of the plate is accomplished. The experimental results demonstrate that, this method enjoys the advantages of being simple and less demanding in image quality and frame of license plate, and makes the correction of the plate easier and more precise compare to some other method such as Hough transform.

12 October 2024
Verified
Conference Paper
Open access
SCShi Cheng
Shi Cheng
ZCZhiqiang Chen
Zhiqiang Chen
JWJing Wang
Jing Wang
This article aims to explore the reform of cooperative education and talent development models between universities and enterprises in the field of computer science, to meet current industry demands and students' developmental needs. An analysis of the current situation reveals some issues in these cooperative models that require reform. Therefore, this paper proposes a series of reform suggestions, including strengthening university-enterprise cooperation, innovating in teaching and talent development models, transforming the roles of teachers and students, and improving the evaluation mechanism. These suggestions are intended to enhance the comprehensive quality and practical skills of students in computer science, promote the integration of industry and education, and drive the continuous development of computer science education. The measures discussed can provide references for relevant educational institutions and enterprises to promote ongoing innovation and improvement in the cooperative education and talent development models in computer science.