Search

Your search keyword '"Wu, Junfei"' showing total 122 results
Start Over Author "Wu, Junfei" Publication Year Range Last 10 years
122 results on '"Wu, Junfei"'

1. MME-RealWorld: Could Your Multimodal LLM Challenge High-Resolution Real-World Scenarios that are Difficult for Humans?

Author

Zhang, Yi-Fan, Zhang, Huanyu, Tian, Haochen, Fu, Chaoyou, Zhang, Shuangqing, Wu, Junfei, Li, Feng, Wang, Kun, Wen, Qingsong, Zhang, Zhang, Wang, Liang, Jin, Rong, and Tan, Tieniu

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Comprehensive evaluation of Multimodal Large Language Models (MLLMs) has recently garnered widespread attention in the research community. However, we observe that existing benchmarks present several common barriers that make it difficult to measure the significant challenges that models face in the real world, including: 1) small data scale leads to a large performance variance; 2) reliance on model-based annotations results in restricted data quality; 3) insufficient task difficulty, especially caused by the limited image resolution. To tackle these issues, we introduce MME-RealWorld. Specifically, we collect more than $300$K images from public datasets and the Internet, filtering $13,366$ high-quality images for annotation. This involves the efforts of professional $25$ annotators and $7$ experts in MLLMs, contributing to $29,429$ question-answer pairs that cover $43$ subtasks across $5$ real-world scenarios, extremely challenging even for humans. As far as we know, MME-RealWorld is the largest manually annotated benchmark to date, featuring the highest resolution and a targeted focus on real-world applications. We further conduct a thorough evaluation involving $28$ prominent MLLMs, such as GPT-4o, Gemini 1.5 Pro, and Claude 3.5 Sonnet. Our results show that even the most advanced models struggle with our benchmarks, where none of them reach $60\%$ accuracy. The challenges of perceiving high-resolution images and understanding complex real-world scenarios remain urgent issues to be addressed. The data and evaluation code are released at https://mme-realworld.github.io/ ., Comment: Project Page: https://mme-realworld.github.io/

Published
2024

2. Logical Closed Loop: Uncovering Object Hallucinations in Large Vision-Language Models

Author

Wu, Junfei, Liu, Qiang, Wang, Ding, Zhang, Jinghao, Wu, Shu, Wang, Liang, and Tan, Tieniu

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Artificial Intelligence, Computer Science - Computation and Language, Computer Science - Machine Learning
Abstract

Object hallucination has been an Achilles' heel which hinders the broader applications of large vision-language models (LVLMs). Object hallucination refers to the phenomenon that the LVLMs claim non-existent objects in the image. To mitigate the object hallucinations, instruction tuning and external model-based detection methods have been proposed, which either require large-scare computational resources or depend on the detection result of external models. However, there remains an under-explored field to utilize the LVLM itself to alleviate object hallucinations. In this work, we adopt the intuition that the LVLM tends to respond logically consistently for existent objects but inconsistently for hallucinated objects. Therefore, we propose a Logical Closed Loop-based framework for Object Hallucination Detection and Mitigation, namely LogicCheckGPT. In specific, we devise logical consistency probing to raise questions with logical correlations, inquiring about attributes from objects and vice versa. Whether their responses can form a logical closed loop serves as an indicator of object hallucination. As a plug-and-play method, it can be seamlessly applied to all existing LVLMs. Comprehensive experiments conducted on three benchmarks across four LVLMs have demonstrated significant improvements brought by our method, indicating its effectiveness and generality., Comment: Accept to ACL 2024; 19 Pages, 15 Figures, 6 Tables

Published
2024

3. Can Large Language Models Detect Rumors on Social Media?

Author

Liu, Qiang, Tao, Xiang, Wu, Junfei, Wu, Shu, and Wang, Liang

Subjects
Computer Science - Information Retrieval, Computer Science - Computation and Language
Abstract

In this work, we investigate to use Large Language Models (LLMs) for rumor detection on social media. However, it is challenging for LLMs to reason over the entire propagation information on social media, which contains news contents and numerous comments, due to LLMs may not concentrate on key clues in the complex propagation information, and have trouble in reasoning when facing massive and redundant information. Accordingly, we propose an LLM-empowered Rumor Detection (LeRuD) approach, in which we design prompts to teach LLMs to reason over important clues in news and comments, and divide the entire propagation information into a Chain-of-Propagation for reducing LLMs' burden. We conduct extensive experiments on the Twitter and Weibo datasets, and LeRuD outperforms several state-of-the-art rumor detection models by 3.2% to 7.7%. Meanwhile, by applying LLMs, LeRuD requires no data for training, and thus shows more promising rumor detection ability in few-shot or zero-shot scenarios.

Published
2024

4. Out-of-distribution Evidence-aware Fake News Detection via Dual Adversarial Debiasing

Author

Liu, Qiang, Wu, Junfei, Wu, Shu, and Wang, Liang

Subjects
Computer Science - Computation and Language, Computer Science - Artificial Intelligence
Abstract

Evidence-aware fake news detection aims to conduct reasoning between news and evidence, which is retrieved based on news content, to find uniformity or inconsistency. However, we find evidence-aware detection models suffer from biases, i.e., spurious correlations between news/evidence contents and true/fake news labels, and are hard to be generalized to Out-Of-Distribution (OOD) situations. To deal with this, we propose a novel Dual Adversarial Learning (DAL) approach. We incorporate news-aspect and evidence-aspect debiasing discriminators, whose targets are both true/fake news labels, in DAL. Then, DAL reversely optimizes news-aspect and evidence-aspect debiasing discriminators to mitigate the impact of news and evidence content biases. At the same time, DAL also optimizes the main fake news predictor, so that the news-evidence interaction module can be learned. This process allows us to teach evidence-aware fake news detection models to better conduct news-evidence reasoning, and minimize the impact of content biases. To be noted, our proposed DAL approach is a plug-and-play module that works well with existing backbones. We conduct comprehensive experiments under two OOD settings, and plug DAL in four evidence-aware fake news detection backbones. Results demonstrate that, DAL significantly and stably outperforms the original backbones and some competitive debiasing methods.

Published
2023

5. Adversarial Contrastive Learning for Evidence-aware Fake News Detection with Graph Neural Networks

Author

Wu, Junfei, Xu, Weizhi, Liu, Qiang, Wu, Shu, and Wang, Liang

Subjects
Computer Science - Computation and Language, Computer Science - Artificial Intelligence
Abstract

The prevalence and perniciousness of fake news have been a critical issue on the Internet, which stimulates the development of automatic fake news detection in turn. In this paper, we focus on evidence-based fake news detection, where several evidences are utilized to probe the veracity of news (i.e., a claim). Most previous methods first employ sequential models to embed the semantic information and then capture the claim-evidence interaction based on attention mechanisms. Despite their effectiveness, they still suffer from three weaknesses. Firstly, sequential models fail to integrate the relevant information that is scattered far apart in evidences. Secondly, they underestimate much redundant information in evidences may be useless or harmful. Thirdly, insufficient data utilization limits the separability and reliability of representations captured by the model. To solve these problems, we propose a unified Graph-based sEmantic structure mining framework with ConTRAstive Learning, namely GETRAL in short. Specifically, we first model claims and evidences as graph-structured data to capture the long-distance semantic dependency. Consequently, we reduce information redundancy by performing graph structure learning. Then the fine-grained semantic representations are fed into the claim-evidence interaction module for predictions. Finally, an adversarial contrastive learning module is applied to make full use of data and strengthen representation learning. Comprehensive experiments have demonstrated the superiority of GETRAL over the state-of-the-arts and validated the efficacy of semantic mining with graph structure and contrastive learning., Comment: 12 pages; in submission to IEEE TKDE. arXiv admin note: substantial text overlap with arXiv:2201.06885

Published
2022

6. Evidence-aware Fake News Detection with Graph Neural Networks

Author

Xu, Weizhi, Wu, Junfei, Liu, Qiang, Wu, Shu, and Wang, Liang

Subjects
Computer Science - Computation and Language, Computer Science - Artificial Intelligence
Abstract

The prevalence and perniciousness of fake news has been a critical issue on the Internet, which stimulates the development of automatic fake news detection in turn. In this paper, we focus on the evidence-based fake news detection, where several evidences are utilized to probe the veracity of news (i.e., a claim). Most previous methods first employ sequential models to embed the semantic information and then capture the claim-evidence interaction based on different attention mechanisms. Despite their effectiveness, they still suffer from two main weaknesses. Firstly, due to the inherent drawbacks of sequential models, they fail to integrate the relevant information that is scattered far apart in evidences for veracity checking. Secondly, they neglect much redundant information contained in evidences that may be useless or even harmful. To solve these problems, we propose a unified Graph-based sEmantic sTructure mining framework, namely GET in short. Specifically, different from the existing work that treats claims and evidences as sequences, we model them as graph-structured data and capture the long-distance semantic dependency among dispersed relevant snippets via neighborhood propagation. After obtaining contextual semantic information, our model reduces information redundancy by performing graph structure learning. Finally, the fine-grained semantic representations are fed into the downstream claim-evidence interaction module for predictions. Comprehensive experiments have demonstrated the superiority of GET over the state-of-the-arts., Comment: Accepted by TheWebConf 2022

Published
2022

8. Mechanisms of the TGF-[beta]1/Smad3-signaling pathway in gender differences in alcoholic liver fibrosis

Author

Hong, Xiaomin, Li, Sanqiang, Luo, Renli, Yang, Mengli, Wu, Junfei, Chen, Shuning, and Zhu, Siyu

Subjects
Estradiol, Fibrosis, Collagen, Tumor necrosis factor, Transforming growth factors, Liver diseases, Testosterone, Psychology and mental health
Abstract

The TGF-[beta]1/Smad3-signaling pathway and gender differences were investigated in alcoholic liver fibrosis. Mice were divided into female normal, female model, male normal, and male model groups. Liver injury and fibrosis were assessed using histopathology and serology. Western blotting was performed to analyze the expression of relevant factors. HSC-T6 cells were divided into estradiol + saline, estradiol + ethanol, testosterone + saline, and testosterone + ethanol groups, and similar assessments were conducted in vitro. Compared with the female model group, the male model group exhibited significantly increased GPT, GOT, TNF-[alpha], IL-6, and testosterone levels, fibrosis rate, and TGF-[beta]1, Smad3, and PCNA expression, and significantly decreased estradiol levels and Caspase-3 expression. The apoptosis rate was higher in the estradiol + ethanol group than in the testosterone + ethanol group, although the testosterone + ethanol group exhibited significantly increased TNF-[alpha], IL-6, Collagen-I, [alpha]-SMA, TGF-[beta]1, Smad3, and PCNA expression, and significantly decreased Caspase-3 expression. Alcoholic liver fibrosis showed significant gender differences associated with the TGF-[beta]1/Smad3-signaling pathway. Keywords: Alcoholic liver fibrosis, Gender differences, Estradiol, Testosterone, TGF-[beta]1/Smad3-signaling pathway, Author(s): Xiaomin Hong[sup.1,2], Sanqiang Li[sup.1,2], Renli Luo[sup.1,2], Mengli Yang[sup.1,2], Junfei Wu[sup.1,2], Shuning Chen[sup.3] and Siyu Zhu[sup.1,2] Background Alcoholic liver fibrosis results from the impact of ethanol and its metabolites on [...]

Published
2024
Full Text
View/download PDF

10. Geochronology and geochemistry of Early Cretaceous granitic plutons in northern Great Xing’an Range, NE China, and implications for geodynamic setting

Author

Luo Niangang, Gao Lianfeng, Zhang Jing, Zhang Zhenguo, Wu Junfei, Cui Jianyu, and Xing Jie

Subjects
great xing’an range, early cretaceous, i-type granite, mongol-okhotsk tectonic regime, paleo-pacific tectonic regime, Geology, QE1-996.5
Abstract

Early Cretaceous granitic rocks are widely distributed in the Great Xing’an Range, northeast China. However, their precise age and tectonic evolution remain controversial. This study presents new zircon U–Pb, Lu–Hf, and whole-rock geochemical data for the granitic plutons in the Yili area, Xing’an Massif, northern Great Xing’an Range. The aim of this study was to decipher the tectono-thermal history and obtain further understanding of the geodynamic setting of the large-scale Late Mesozoic magmatism in the Great Xing’an Range. Zircon U–Pb age dating indicated that the plutons were emplaced during the Early Cretaceous, with ages of 132.2–130.1 Ma. The plutons showed similar geochemical features, characterized by high concentrations of SiO2 and Na2O + K2O and low concentrations of P2O5, CaO, MgO, and TFe2O3. The plutons were enriched in light rare earth elements and large ion lithophile elements and depleted in heavy earth elements and high-field-strength elements. Such features indicate that the plutons are highly fractionated with I-type granite affinity. These findings, combined with pluton ε Hf(t) values of + 5.25 to + 8.28 and two-stage model ages (t DM2) of 661–855 Ma, indicate that the primary magmas originated from partial melting of juvenile basaltic crustal material accreted during the Neoproterozoic. These results combined with regional geological evolution indicated that the generation of Early Cretaceous plutons in the northern Great Xing’an Range might be closely related to the post-collisional gravitational collapse and subsequent extension resulting from the closure of the Mongolia-Okhotsk Ocean. Their generation also superimposed the back-arc extension resulting from retraction of the Paleo-Pacific subduction plate.

Published
2022
Full Text
View/download PDF

12. Drag Reduction Mechanism of Array Microslit-Opposed Jets in the Passivation Leading Edge of High-Speed Aircrafts.

Author

Zhao, Yongsheng, Zhang, Jiang, Dong, Jingang, and Wu, Junfei

Subjects
DRAG reduction, DRAG (Aerodynamics), PASSIVATION, MACH number, MODEL airplanes, RADIAL basis functions
Abstract

In order to meet the demanding drag reduction requirements of high-speed aircraft, the drag reduction mechanism of array microslit-opposed jets in the passivation leading edge is studied using a numerical method based on the shear stress transfer (SST) k-ω two-path turbulence model. The computations are performed with the commercial CFD solver FLUENT 13.0 code. The interference characteristics of the array jets are analyzed, the influence laws of the microslit array mode and the angle of attack on drag reduction performance are explored, and a prediction method of passivation leading edge drag characteristics based on radial basis function (RBF) neural network is constructed. The research conditions are as follows: the incoming Mach number is 2.0, the angle of attack ranges from 0° to 30°, and the number of microslits is 0 to 5. The results show that when there is interference between microslit jets, the interaction between adjacent reverse reflux regions will change the shape of the reflux regions. When there is a microslit jet in the center line, the position of the detached shock wave is more affected by the angle of attack. The drag characteristics of the leading edge are determined by the location of the detached shock wave and the characteristics of the reflux regions. The mode of microslits determines the drag reduction efficiency and sensitivity to the angle of attack. There may be contradictions in the design of array microslit-opposed jet schemes regarding the overall drag reduction efficiency, local load, and drag reduction stability. The drag prediction model based on the RBF neural network has an accuracy of 5%, which can provide support for the design of array microslit-opposed jets. Practical Applications: The array microslit-opposed jets provide a new technical approach for reducing drag on the passivation leading edge. The paper studies its flow characteristics and laws and provides relevant conclusions. It can serve as a theoretical reference during design. The number of microslits is not the more the better. When designing microslit patterns, many factors need to be considered. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

17. Quantitative Relationship Between Fluorescence Intensity and Equivalence Ratio of Kerosene

Author

Zhao, Yongsheng, Wu, Junfei, Tian, Xiaohu, Lin, Yuzhen, Wei, Wei, Angrisani, Leopoldo, Series Editor, Arteaga, Marco, Series Editor, Panigrahi, Bijaya Ketan, Series Editor, Chakraborty, Samarjit, Series Editor, Chen, Jiming, Series Editor, Chen, Shanben, Series Editor, Chen, Tan Kay, Series Editor, Dillmann, Rüdiger, Series Editor, Duan, Haibin, Series Editor, Ferrari, Gianluigi, Series Editor, Ferre, Manuel, Series Editor, Hirche, Sandra, Series Editor, Jabbari, Faryar, Series Editor, Jia, Limin, Series Editor, Kacprzyk, Janusz, Series Editor, Khamis, Alaa, Series Editor, Kroeger, Torsten, Series Editor, Liang, Qilian, Series Editor, Martin, Ferran, Series Editor, Ming, Tan Cher, Series Editor, Minker, Wolfgang, Series Editor, Misra, Pradeep, Series Editor, Möller, Sebastian, Series Editor, Mukhopadhyay, Subhas, Series Editor, Ning, Cun-Zheng, Series Editor, Nishida, Toyoaki, Series Editor, Pascucci, Federica, Series Editor, Qin, Yong, Series Editor, Seng, Gan Woon, Series Editor, Speidel, Joachim, Series Editor, Veiga, Germano, Series Editor, Wu, Haitao, Series Editor, Zhang, Junjie James, Series Editor, and Zhang, Xinguo, editor

Published
2019
Full Text
View/download PDF

25. Mechanisms of the TGF-β1/Smad3-signaling pathway in gender differences in alcoholic liver fibrosis.

Author

Hong, Xiaomin, Li, Sanqiang, Luo, Renli, Yang, Mengli, Wu, Junfei, Chen, Shuning, and Zhu, Siyu

Abstract

The TGF-β1/Smad3-signaling pathway and gender differences were investigated in alcoholic liver fibrosis. Mice were divided into female normal, female model, male normal, and male model groups. Liver injury and fibrosis were assessed using histopathology and serology. Western blotting was performed to analyze the expression of relevant factors. HSC-T6 cells were divided into estradiol + saline, estradiol + ethanol, testosterone + saline, and testosterone + ethanol groups, and similar assessments were conducted in vitro. Compared with the female model group, the male model group exhibited significantly increased GPT, GOT, TNF-α, IL-6, and testosterone levels, fibrosis rate, and TGF-β1, Smad3, and PCNA expression, and significantly decreased estradiol levels and Caspase-3 expression. The apoptosis rate was higher in the estradiol + ethanol group than in the testosterone + ethanol group, although the testosterone + ethanol group exhibited significantly increased TNF-α, IL-6, Collagen-I, α-SMA, TGF-β1, Smad3, and PCNA expression, and significantly decreased Caspase-3 expression. Alcoholic liver fibrosis showed significant gender differences associated with the TGF-β1/Smad3-signaling pathway. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

26. DESIGN AND APPLICATION OF NEW TYPE OF CRANK AUTOMATIC ADJUSTMENT MECHANISM OF PUMPING UNIT

Author

WU JunFei, WANG Qian, JIN Xin, and XIAO Shu

Subjects
Pumping unit, Crank balance, Structural design, Theoretical calculation, Finite element analysis, Field test, Mechanical engineering and machinery, TJ1-1570, Materials of engineering and construction. Mechanics of materials, TA401-492
Abstract

To lower the labor intensity, reduce energy consumption, improve efficiency, increase the service life of the pumping unit and achieve automatic adjustment of pumping unit, based on the CYJ10-4.2-53 HF type pumping unit, this paper completes the structure and parameter design of the crank automatic balance device. The key transmission part screw is selected and checked based on the fourth strength theory. ANSYS Workbench is used to analyze the finite element strength. It is concluded that the theoretical check is basically as the same as the simulation result, the theoretical calculation is feasible, and the overall structure meets the strength requirements.Finally, the pumping unit is tested in the field, the instrument measured its balance meets the requirements, the energy saving effect is satisfactory, and it has a certain promotion and application prospects.

Published
2019
Full Text
View/download PDF

27. The Pivotal Radical Intermediate [Au21(SR)15]+in the Ligand-Exchange-Induced Size-Reduction of [Au23(SR)16]−to Au16(SR)12

Author

Cui, Mengting, Shi, Yanan, Ma, Xiangyu, Li, Qingliang, Chen, Ling, Zhang, Lichao, Wu, Junfei, Yu, Haizhu, and Zhu, Manzhou

Abstract

The atomic precision of sub-nanometer-sized metal nanoclusters makes it possible to elucidate the kinetics of metal nanomaterials from the molecular level. Herein, the size reduction of an atomically precise [Au23(CHT)16]−(HCHT = cyclohexanethiol) cluster upon ligand exchange with HSAdm (1-adamantanethiol) has been reported. During the 16 h conversion of [Au23(CHT)16]−to Au16(SR)12, the neutral 6e Au21(SR)15, and its 1e-reduction state, i.e. the 5e, cationic radical, [Au21(SR)15]+, are active intermediates to account for the formation of thermodynamically stable Au16products. The combination of spectroscopic monitoring (with UV–vis and ESI-MS) and DFT calculations indicates the preferential size-reduction on the corner Au atoms on the core surface and the terminal Au atoms on longer AunSn+1staples. This study provides a reassessment on the electronic state of the Au21structure and highlights the single electron transfer processes in cluster systems and thus the importance of the EPR analysis on the mechanistic issues.

Published
2024
Full Text
View/download PDF

38. Nicotinamide adenine dinucleotide (NAD+) reduction enabled by an atomically precise Au-Ag alloy nanocluster.

Author

Chen, Ling, Du, Yonglei, Lv, Ying, Fan, Daoqing, Wu, Junfei, Wu, Lingbao, Cui, Mengting, Yu, Haizhu, and Zhu, Manzhou

Subjects
NAD (Coenzyme), ELECTROSPRAY ionization mass spectrometry, NICOTINAMIDE, ADENINE, ALLOYS, HYDROXYL group
Abstract

The redox property of the ultrasmall coinage nanoclusters (with several to tens of Au/Ag atoms) has elucidated the electron-transfer capacity of nanoclusters, and has been successfully utilized in a variety of redox conversions (such as from CO2 to CO). Nevertheless, their biological applications are mainly restricted by the scarcity of atomically precise, water-soluble metal nanoclusters, and the limited application (mainly on the decomposition of H2O2 in these days). Herein, mercaptosuccinic acid (MSA) protected ultrasmall alloy AuAg nanoclusters were prepared, and the main product was determined [Au3Ag5(MSA)3]− by electrospray ionization mass spectrometry (ESI-MS). The clusters can not only mediate the decomposition of H2O2 to generate hydroxyl radicals, but is also able to mediate the reduction of nicotinamide adenine dinucleotide (NAD) to its reduced form of NADH. This is the first time that the atomically precise metal nanoclusters were used to mediate the coenzyme reduction. The preliminary mechanistic insights imply the reaction to be driven by the hydrogen bonding between the carboxylic groups (on the surface of MSA) and the amino N—H bonds (on NAD). In this context, the presence of the carboxylic groups, the sub-nanometer size regime (∼ 1 nm), and the synergistic effect of the Au-Ag clusters are pre-requisite to the NAD reduction. [ABSTRACT FROM AUTHOR]

Published
2023
Full Text
View/download PDF

41. Atomization Characteristics of Kerosene in Crossflow with an Incident Shock Wave.

Author

Zhao, Yongsheng, Wu, Junfei, and Mu, Xiangyang

Subjects
SHOCK waves, KELVIN-Helmholtz instability, ATOMIZATION, KEROSENE, JETS (Fluid dynamics), LASER peening, COLLISIONLESS plasmas, PARTICLE size distribution
Abstract

An unsteady numerical simulation method was used in order to explore more efficient atomization methods for liquid fuel in scramjet combustors and to study the influence of different shock wave incident positions on the atomization characteristics of kerosene in crossflow. The wedge compression surface was used to generate the incident shock wave, and the incident position of the shock wave on the fuel jet was controlled by changing the angle of the wedge surface. The inlet Mach number was 2.01; the total temperature was 300 K, and the momentum ratio was 12. The research results show that as the incident position of the shock wave moves upstream, the penetration depth of the jet is essentially unchanged, but the inner edge trajectory of the jet is closer to the wall. Because the shock wave affects the Kelvin–Helmholtz instability of the jet, the unsteadiness of the jet root is strengthened, and the unsteadiness downstream of the jet is weakened. The atomization of the jet and the stability of the particle-size distribution are, thus, realized more quickly. The incident shock wave reduces the Sauter mean diameter of the jet section and makes the droplet distribution more uniform. The incident shock wave makes the atomization angle of the jet along the flow direction increase first and then decrease. The changes in the jet characteristics are determined by the changes in the reflux region, momentum transport, and pressure distribution caused by the incident shock wave. [ABSTRACT FROM AUTHOR]

Published
2023
Full Text
View/download PDF

46. miR-133b Suppresses Invasion and Migration of Gastric Cancer Cells via the COL1A1/TGF-β Axis

Author

Guo, Yuan, Lu, Guochun, Mao, Huahui, Zhou, Shengkun, Tong, Xiangmei, Wu, Junfei, Sun, Qiang, Xu, Hui, and Fang, Fu

Subjects
COL1A1, gastric cancer, invasion and migration, TGF-β signaling pathway, Original Research, miR-133b
Abstract

Objective The study aimed to explore the mechanism of miR-133b regulating the invasion and migration of gastric cancer (GC) cells via the COL1A1/TGF-β axis. Methods The miRNA expression profiles of GC downloaded from TCGA database were subjected to differential analysis to determine the target miRNA of interest, and the target genes of the miRNA were predicted by bioinformatics. GSEA was used for gene enrichment analysis. qRT-PCR was carried out to detect gene expression in GC cells. The effect of miR-133b on GC cells was examined by CCK-8, wound healing and Transwell assays. Western blot was conducted to assess the protein expression of EMT-related proteins. The binding relationship between genes was verified by dual-luciferase reporter gene assay. Results The expression of miR-133b was markedly downregulated in GC tissue, while that of COL1A1 was upregulated. Overexpression of miR-133b decreased the migration and invasion of GC cells, and the EMT process was inhibited as well, while inverse results were observed when miR-133b was silenced. COL1A1 was a target gene of miR-133b and its overexpression had a significant impact on the prognosis of patients. GSEA pathway enrichment results showed that COL1A1 was markedly enriched in the TGF-β signaling pathway. In addition, COL1A1 overexpression induced the activation of the TGF-β signaling pathway to promote proliferation and migration of GC cells, whereas miR-133b overexpression suppressed the signaling pathway. Thus, overexpression of miR-133b and COL1A1 simultaneously would reverse the inhibitory effect of miR-133b on cell invasion and migration. Conclusion In this study, miR-133b was found to inhibit the invasion and migration of GC cells via the COL1A1/TGF-β axis, which provides a new research direction for the diagnosis and targeted therapy of GC.

Published
2020

47. A compound control method for coagulant dosing in Waterworks

Author

Wu Junfei

Subjects
Water flow, Sedimentation (water treatment), Environmental science, Coagulation (water treatment), Process control, Water quality, Raw water, Turbidity, Pulp and paper industry, Effluent
Abstract

Changes of raw water quality and water flow has obvious influence on the coagulation process of waterworks. However, most waterworks still adopt manual control method based on fixed proportion of water flow. The coagulant dosage can not be adjusted in time according to the change of raw water quality and water flow, so that the coagulant dosage can not be adjusted in time. It is easy to cause large fluctuation of turbidity of effluent from sedimentation tank with large lag. In order to improve the self-adaptive ability of coagulation process and stabilize the turbidity of effluent from sedimentation tank, a composite control method of coagulant automatic dosing based on RBF neural network (RBFNN) and PI control is studied in this paper. RBFNN provides timely feed-forward control information for coagulant automatic dosing according to raw water quality. PI control is used for feedback control of coagulant automatic dosing. The results show that, compared with manual control, the proposed compound control has better adaptability to the changes of raw water quality and water flow, and the turbidity of the effluent of sedimentation tank is more stable.

Published
2021

50. A Distributed Video Monitoring System Architecture for Semantic Retrieval

Author

Wang Lu, Li Qingdang, Wu Junfei, and Sun Zhen

Subjects
Source data, Computer architecture, Computer science, Server, Distributed data store, Search engine indexing, Systems architecture, Video monitoring, Architecture, Field (computer science)
Abstract

The storage and intelligent retrieval of massive video data is a hot topic in the field of surveillance. This paper designs a system architecture for distributed storage, indexing, and querying of ultra-magnitude video and its source data. By adopting a loosely coupled architecture, the various processing modules and services are completely separated. The system supports full processing from storage to semantic retrieval, enabling seamless expansion of the system and supporting uninterrupted video streaming.

Published
2019

Catalog

Books, media, physical & digital resources
See catalog results