Rong Shan

Kunming University of Science and Technology

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Ruiming Tang

Huawei Technologies (China)

Aiguang Wang

Shandong Provincial QianFoShan Hospital

Guanghe Dong

Kunming University of Science and Technology

Yong Yu

Shanghai Jiao Tong University

Jianghao Lin

Guangdong University of Technology

Bo Chen

Huawei Technologies (China)

Yun Zeng

Kunming University of Science and Technology

Xinyi Dai

Ministry of Education

Jigang Zhao

East China University of Science and Technology

Juan Duan

Kunming University of Science and Technology

Cooperative Institutions

Shanghai University of Traditional Chinese Medicine

Shanghai Jiao Tong University

Yueyang Hospital

Central South University

Shandong University

Xiangya Hospital Central South University

Chinese Academy of Sciences

Chinese Academy of Medical Sciences & Peking Union Medical College

Shandong Provincial QianFoShan Hospital

Shandong First Medical University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Large Language Models Make Sample-Efficient Recommender Systems

arXiv (Cornell University) (2024)

Jianghao Lin Xinyi Dai Rong Shan Bo Chen Ruiming Tang

Large language models (LLMs) have achieved remarkable progress in the field of natural language processing (NLP), demonstrating remarkable abilities in producing text that resembles human language for various tasks. This opens up new opportunities for employing them in recommender systems (RSs). In this paper, we specifically examine the sample efficiency of LLM-enhanced recommender systems, which pertains to the model's capacity to attain superior performance with a limited quantity of training data. Conventional recommendation models (CRMs) often need a large amount of training data because of the sparsity of features and interactions. Hence, we propose and verify our core viewpoint: Large Language Models Make Sample-Efficient Recommender Systems. We propose a simple yet effective framework (i.e., Laser) to validate the viewpoint from two aspects: (1) LLMs themselves are sample-efficient recommenders; and (2) LLMs, as feature generators and encoders, make CRMs more sample-efficient. Extensive experiments on two public datasets show that Laser requires only a small fraction of training samples to match or even surpass CRMs that are trained on the entire training set, demonstrating superior sample efficiency.

Sample (material)

10.48550/arxiv.2406.02368

Cite

Citations (0)

Role of autophagy-related genes in liver cancer prognosis

Genomics (2024)

Yuling Zhou Rong Shan Wangti Xie Qiang Zhou Qinghua Yin

Autophagy, a highly conserved process of protein and organelle degradation, has emerged as a critical regulator in various diseases, including cancer progression. In the context of liver cancer, the predictive value of autophagy-related genes remains ambiguous. Leveraging chip datasets from the TCGA and GTEx databases, we identified 23 differentially expressed autophagy-related genes in liver cancer. Notably, five key autophagy genes, PRKAA2, BIRC5, MAPT, IGF1, and SPNS1, were highlighted as potential prognostic markers, with MAPT showing significant overexpression in clinical samples. In vitro cellular assays further demonstrated that MAPT promotes liver cancer cell proliferation, migration, and invasion by inhibiting autophagy and suppressing apoptosis. Subsequent in vivo studies further corroborated the pro-tumorigenic role of MAPT by suppressing autophagy. Collectively, our model based on the five key genes provides a promising tool for predicting liver cancer prognosis, with MAPT emerging as a pivotal factor in tumor progression through autophagy modulation.

Liver Cancer

10.1016/j.ygeno.2024.110852

Cite

Citations (1)

Progress in the knowledge on the transformation of lung adenocarcinoma to small-cell lung cancer

Journal of Cancer Research and Therapeutics (2023)

Aiguang Wang Cuiping Han Huijun Zhao Zhaomin Zheng Xin Ye

ABSTRACT Lung cancer is a common type of carcinoma and is the leading cause of cancer-related deaths worldwide. The two broad histological subtypes of lung cancer are non-small-cell lung cancer (NSCLC), which accounts for 85% of cases and includes adenocarcinoma and squamous cell carcinoma, and small-cell lung cancer (SCLC), which accounts for 15% of cases. Substantial improvements in treatment have led to remarkable progress and changed outcomes for many patients in the past two decades. However, with prolonged survival time and awareness of repeat biopsy, more and more patients with lung cancer have been found to undergo a histological transformation during treatment, with lung adenocarcinoma (LAdC) to SCLC transformation being the most frequent. In this article, we summarized findings on the mechanism, clinical characteristics, therapeutic strategies, and predictors of the transformation of LAdC to SCLC. A non-systematic narrative review was performed using the Pubmed/MEDLINE (US National Library of Medicine National Institutes of Health) database with the following keywords: “transformation from NSCLC to SCLC,” “transformation from lung adenocarcinoma to small-cell lung cancer,” “NSCLC transformation in SCLC,” and “NSCLC and transformation and SCLC.” Articles published until June 2022 were analyzed. Search results were limited to human studies without restriction for language.

10.4103/jcrt.jcrt_1842_22

Cite

Citations (3)

Study on the Thermal Field of a Hydro-Generator under the Effect of a Plateau Climate

Energies (2024)

Rong Shan Juan Duan Yun Zeng Jing Qian Guanghe Dong

With the advancement in the developmental project on the lower reaches of the Yarlung Tsangpo River, the influence of the plateau climate environment on the performance of a hydro-generator has received more and more attention from researchers. This study numerically simulated the thermal field of a hydro-generator with a 20 MW capacity under the effect of a plateau climate. Ambient pressure and temperature are two main factors that affect the temperature distribution of the generator. In addition, temperature distributions with different speeds are also studied under a plateau climate. The results show that the generator temperature decreases with increasing air pressure and speed. The generator temperature increases linearly with increasing ambient temperature. Among them, when the pressure lies in the range of 25–85 kPa, the temperature change gradient of the stator structure is very large. The temperature difference gradually decreases with the increase in air pressure. The temperature gradient gradually slows down when the air pressure is above 85 kPa. When the pressure is located at 55–85 kPa, the average temperature difference of the stator windings is 6.325 °C, and the average temperature difference of the stator core is 3.815 °C. Finally, the temperature distribution pattern can provide a basis for staff in different barometric pressure regions. It can also improve the safety and reliability of the hydro-generator under the effect of a plateau climate, which is important for improving its integrated hydraulic performance.

10.3390/en17040932

Cite

Citations (0)

p型ホイスラー化合物:MgO(100)上にエピタキシャル成長させたNiYBi薄膜の成長,構造,および特性

Applied Physics Letters (2012)

Rong Shan Siham Ouardi H Fecher Gerhard Li Gao Kellock Andrew

Source

Cite

Citations (5)

High-efficiency and High-quality Cultivation Techniques of Cirtrs grandis

Jiangxi Forestry Science and Technology (2010)

Rong Shan

High-yield and high-quality cultivation and management techni ques of characteristic varieties of local Citrus grandis named Red Citrus grandiswere explored in this paper,it emphasized on practical experiences of high-standard planning garden,planting conditions,cultivation of high-yield tree types,pest control,frost prevention and fruit harvest,and so on,in order to provide technical supports for the implementation of the industrialization of Citrus grandis.

Frost (temperature)

Industrialisation

Source

Cite

Citations (0)

Full-Stack Optimized Large Language Models for Lifelong Sequential Behavior Comprehension in Recommendation

arXiv (Cornell University) (2025)

Rong Shan Jiachen Zhu Jianghao Lin Chenxu Zhu Bo Chen

In this paper, we address the lifelong sequential behavior incomprehension problem in large language models (LLMs) for recommendation, where LLMs struggle to extract useful information from long user behavior sequences, even within their context limits. To tackle this, we propose ReLLaX (Retrieval-enhanced Large Language models Plus), a framework offering optimization across data, prompt, and parameter levels. At the data level, we introduce Semantic User Behavior Retrieval (SUBR) to reduce sequence heterogeneity, making it easier for LLMs to extract key information. For prompt-level enhancement, we employ Soft Prompt Augmentation (SPA) to inject collaborative knowledge, aligning item representations with recommendation tasks and improving LLMs's exploration of item relationships. Finally, at the parameter level, we propose Component Fully-interactive LoRA (CFLoRA), which enhances LoRA's expressiveness by enabling interactions between its components, allowing better capture of sequential information. Moreover, we present new perspectives to compare current LoRA-based LLM4Rec methods, i.e. from both a composite and a decomposed view. We theoretically demonstrate that the ways they employ LoRA for recommendation are degraded versions of our CFLoRA, with different constraints on atom component interactions. Extensive experiments on three public datasets demonstrate ReLLaX's superiority over existing baselines and its ability to mitigate lifelong sequential behavior incomprehension effectively.

10.48550/arxiv.2501.13344

Cite

Citations (0)

Influencing factors of transient elastography in detecting liver stiffness

Experimental and Therapeutic Medicine (2016)

Rong Shan Hong Yin Wenjuan Yang Jianzhi Li Meifang Zhang

Liver stiffness, which correlates well with liver fibrosis stage, can be measured noninvasively by transient elastography, also known as Fibroscan. The present study aimed to determine the independent factors influencing Fibroscan detection by multiple regression analysis. A total of 181 patients who required liver biopsy were enrolled. Liver stiffness measurement (LSM) was detected by Fibroscan on the day of liver biopsy, while clinical information and routine biochemical examination results were also collected. Correlation was analyzed by Spearman's correlation, and multiple regression analysis was performed to analyze the independent influencing factors. The results demonstrated that platelet (PLT) levels, serum albumin (ALB), prothrombin activity (PTA) and body mass index (BMI) were independent predictors of liver stiffness. The contribution of these four predictors to the regression equation was in the following descending order: PLT (negative correlation) > ALB (negative correlation) > PTA (negative correlation) > BMI (positive correlation). In conclusion, the parameters of PLT, ALB, PTA and BMI are independent predicting factors affecting Fibroscan detection. Therefore, the diagnosis and evaluation of liver fibrosis should comprehensively consider the results of Fibroscan, and clinical and laboratory examinations.

Transient elastography

Prothrombin time

10.3892/etm.2016.3617

Cite

Citations (12)

The Construction of the index system for nurses' innovation competence evaluation in China: A Delphi study

Heliyon (2024)

Lili Ma Yajuan Zhang Rong Shan Jin-Xiu Yao Huiren Zhuang

Delphi Method

Delphi

10.1016/j.heliyon.2024.e38530

Cite