LIVIVO - Search results -

Search results

Result 1 - 10 of total 11

Search options

Book ; Online: Cross-Model Comparative Loss for Enhancing Neuronal Utility in Language Understanding

Zhu, Yunchang / Pang, Liang / Wu, Kangxi / Lan, Yanyan / Shen, Huawei / Cheng, Xueqi

2023

Abstract: Current natural language understanding (NLU) models have been continuously scaling up, both in terms of model size and input context, introducing more hidden and input neurons. While this generally improves performance on average, the extra neurons do ... ...

Abstract	Current natural language understanding (NLU) models have been continuously scaling up, both in terms of model size and input context, introducing more hidden and input neurons. While this generally improves performance on average, the extra neurons do not yield a consistent improvement for all instances. This is because some hidden neurons are redundant, and the noise mixed in input neurons tends to distract the model. Previous work mainly focuses on extrinsically reducing low-utility neurons by additional post- or pre-processing, such as network pruning and context selection, to avoid this problem. Beyond that, can we make the model reduce redundant parameters and suppress input noise by intrinsically enhancing the utility of each neuron? If a model can efficiently utilize neurons, no matter which neurons are ablated (disabled), the ablated submodel should perform no better than the original full model. Based on such a comparison principle between models, we propose a cross-model comparative loss for a broad range of tasks. Comparative loss is essentially a ranking loss on top of the task-specific losses of the full and ablated models, with the expectation that the task-specific loss of the full model is minimal. We demonstrate the universal effectiveness of comparative loss through extensive experiments on 14 datasets from 3 distinct NLU tasks based on 4 widely used pretrained language models, and find it particularly superior for models with few parameters or long input.
Keywords	Computer Science - Computation and Language ; Computer Science - Information Retrieval ; Computer Science - Machine Learning
Subject code	006
Publishing date	2023-01-09
Publishing country	us
Document type	Book ; Online
Database	BASE - Bielefeld Academic Search Engine (life sciences selection)

Full text online

Full text

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Book ; Online: LoL

Zhu, Yunchang / Pang, Liang / Lan, Yanyan / Shen, Huawei / Cheng, Xueqi

A Comparative Regularization Loss over Query Reformulation Losses for Pseudo-Relevance Feedback

2022

Abstract: Pseudo-relevance feedback (PRF) has proven to be an effective query reformulation technique to improve retrieval accuracy. It aims to alleviate the mismatch of linguistic expressions between a query and its potential relevant documents. Existing PRF ... ...

Abstract	Pseudo-relevance feedback (PRF) has proven to be an effective query reformulation technique to improve retrieval accuracy. It aims to alleviate the mismatch of linguistic expressions between a query and its potential relevant documents. Existing PRF methods independently treat revised queries originating from the same query but using different numbers of feedback documents, resulting in severe query drift. Without comparing the effects of two different revisions from the same query, a PRF model may incorrectly focus on the additional irrelevant information increased in the more feedback, and thus reformulate a query that is less effective than the revision using the less feedback. Ideally, if a PRF model can distinguish between irrelevant and relevant information in the feedback, the more feedback documents there are, the better the revised query will be. To bridge this gap, we propose the Loss-over-Loss (LoL) framework to compare the reformulation losses between different revisions of the same query during training. Concretely, we revise an original query multiple times in parallel using different amounts of feedback and compute their reformulation losses. Then, we introduce an additional regularization loss on these reformulation losses to penalize revisions that use more feedback but gain larger losses. With such comparative regularization, the PRF model is expected to learn to suppress the extra increased irrelevant information by comparing the effects of different revised queries. Further, we present a differentiable query reformulation method to implement this framework. This method revises queries in the vector space and directly optimizes the retrieval performance of query vectors, applicable for both sparse and dense retrieval models. Empirical evaluation demonstrates the effectiveness and robustness of our method for two typical sparse and dense retrieval models. Comment: Accepted at SIGIR 2022
Keywords	Computer Science - Information Retrieval ; Computer Science - Computation and Language ; Computer Science - Machine Learning
Subject code	006
Publishing date	2022-04-25
Publishing country	us
Document type	Book ; Online
Database	BASE - Bielefeld Academic Search Engine (life sciences selection)

Full text online

Full text

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Article: MiR-107 Regulates Adipocyte Differentiation and Adipogenesis by Targeting Apolipoprotein C-2 (APOC2) in Bovine

Wei, Xuefeng / Zhao, Xue / Shan, Xinyue / Zhu, Yunchang / Wang, Shuzhe / Chen, Hong / Li, Hui / Ma, Yun

Genes. 2022 Aug. 17, v. 13, no. 8

2022

Abstract: Adipogenesis is a complex and precisely orchestrated process mediated by a series of adipogenic regulatory factors. Recent studies have highlighted the importance of microRNAs (miRNAs) in diverse biological processes, most specifically in regulating cell ...

Abstract	Adipogenesis is a complex and precisely orchestrated process mediated by a series of adipogenic regulatory factors. Recent studies have highlighted the importance of microRNAs (miRNAs) in diverse biological processes, most specifically in regulating cell differentiation and proliferation. However, the mechanisms of miRNAs in adipogenesis are largely unknown. In this study, we found that miR-107 expression was higher in bovine adipose tissue than that in other tissues, and there was a downregulation trend during adipocyte differentiation. To explore the function of miR-107 in adipocyte differentiation, agomiR-107 and antiagomiR-107 were transfected into bovine adipocytes, respectively. Oil Red O staining, CCK-8, EdU assays, RT-qPCR, and Western blotting were performed, and the results showed that overexpressed miR-107 significantly suppressed fat deposition and adipocyte differentiation, while knockdown of miR-107 promoted fat deposition and adipocytes differentiation. In addition, through bioinformatics analysis, luciferase reporter assays, RT-qPCR, and Western blotting, we identified apolipoprotein 2 (APOC2) as a target of miR-107. Transfection of siRNA-APOC2 into adipocytes led to suppression in adipocyte differentiation and proliferation, suggesting a positive role of APOC2 in bovine lipogenesis. In summary, our findings suggested that miR-107 regulates bovine adipocyte differentiation and lipogenesis by directly targeting APOC2, and these results. These theoretical and experimental basis for future clarification of the regulation mechanism of adipocyte differentiation and lipogenesis. Moreover, for the highly conserved among different species, miR-107 may be a potential molecular target to be used for the treatment of lipid-related diseases in the future.
Keywords	adipocytes ; adipogenesis ; adipose tissue ; bioinformatics ; cattle ; lipogenesis ; luciferase ; microRNA ; transfection
Language	English
Dates of publication	2022-0817
Publishing place	Multidisciplinary Digital Publishing Institute
Document type	Article
ZDB-ID	2527218-4
ISSN	2073-4425
ISSN	2073-4425
DOI	10.3390/genes13081467
Database	NAL-Catalogue (AGRICOLA)

Order via subito

This service is chargeable due to the Delivery terms set by subito. Orders including an article and supplementary material will be classified as separate orders. In these cases, fees will be demanded for each order.

Article: Transcriptome profiling of mRNAs in muscle tissue of Pinan cattle and Nanyang cattle

Wei, Xuefeng / Zhu, Yunchang / Zhao, Xue / Zhao, Yadi / Jing, Yujia / Liu, Gege / Wang, Shuzhe / Li, Hui / Ma, Yun

Gene. 2022 May 30, v. 825

2022

Abstract: Mammalian muscle development is regulated by complex gene networks at the molecular level. The revelation of gene regulatory mechanisms is an important basis for the study of muscle development and molecular breeding. To analyze the excellent meat ... ...

Abstract	Mammalian muscle development is regulated by complex gene networks at the molecular level. The revelation of gene regulatory mechanisms is an important basis for the study of muscle development and molecular breeding. To analyze the excellent meat performance of Pinan cattle at the molecular level, we performed high-throughput RNA sequencing to analyze the key regulatory genes that determine the muscle quality traits in Pinan cattle (n = 3) and Nanyang cattle (n = 3). We identified 57 differentially expressed genes in muscle tissue of Pinan cattle compared to that of Nanyang cattle, including 32 upregulated and 25 downregulated genes. GO enrichment analysis showed that these genes were significantly enriched in ‘molecular function’, including voltage-gated ion channel activity, calcium channel activity and calcium ion binding, and KEGG pathway analysis results revealed that adrenergic signaling in cardio myocytes, cell adhesion molecules and inositol phosphate metabolism pathway were significantly enriched. We identified the reliability of RNA-Seq data through RT-qPCR. Meanwhile, we found that GSTA3, PLCB1 and ISYNA1 genes are highly expressed in muscle tissue of Pinan cattle, and these genes play important roles in PI3K/Akt, MEK1/2-ERK and p53-ISYNA1 signaling pathway. In summary, our results suggested that these differentially expressed genes may play important roles in muscle development in Pinan cattle. However, the functions and mechanism of these significantly differential expressed genes should be investigated in future studies.
Keywords	calcium ; calcium channels ; cattle ; cell adhesion ; gene expression regulation ; inositols ; meat ; metabolism ; muscle development ; muscle tissues ; muscles ; myocytes ; phosphates ; transcriptome
Language	English
Dates of publication	2022-0530
Publishing place	Elsevier B.V.
Document type	Article
ZDB-ID	391792-7
ISSN	1879-0038 ; 0378-1119
ISSN (online)	1879-0038
ISSN	0378-1119
DOI	10.1016/j.gene.2022.146435
Database	NAL-Catalogue (AGRICOLA)

Full text online

Accessible to users with ZB MED library card

In stock of ZB MED Bonn / Germany

Z 79/715: Show issues

Order via subito

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Details ▾

Book ; Online: L2R2

Zhu, Yunchang / Pang, Liang / Lan, Yanyan / Cheng, Xueqi

Leveraging Ranking for Abductive Reasoning

2020

Abstract: The abductive natural language inference task ($\alpha$NLI) is proposed to evaluate the abductive reasoning ability of a learning system. In the $\alpha$NLI task, two observations are given and the most plausible hypothesis is asked to pick out from the ... ...

Abstract	The abductive natural language inference task ($\alpha$NLI) is proposed to evaluate the abductive reasoning ability of a learning system. In the $\alpha$NLI task, two observations are given and the most plausible hypothesis is asked to pick out from the candidates. Existing methods simply formulate it as a classification problem, thus a cross-entropy log-loss objective is used during training. However, discriminating true from false does not measure the plausibility of a hypothesis, for all the hypotheses have a chance to happen, only the probabilities are different. To fill this gap, we switch to a ranking perspective that sorts the hypotheses in order of their plausibilities. With this new perspective, a novel $L2R^2$ approach is proposed under the learning-to-rank framework. Firstly, training samples are reorganized into a ranking form, where two observations and their hypotheses are treated as the query and a set of candidate documents respectively. Then, an ESIM model or pre-trained language model, e.g. BERT or RoBERTa, is obtained as the scoring function. Finally, the loss functions for the ranking task can be either pair-wise or list-wise for training. The experimental results on the ART dataset reach the state-of-the-art in the public leaderboard. Comment: SIGIR 2020
Keywords	Computer Science - Information Retrieval ; Computer Science - Computation and Language
Subject code	006
Publishing date	2020-05-22
Publishing country	us
Document type	Book ; Online
Database	BASE - Bielefeld Academic Search Engine (life sciences selection)

Full text online

Full text

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Book ; Online: Adaptive Information Seeking for Open-Domain Question Answering

Zhu, Yunchang / Pang, Liang / Lan, Yanyan / Shen, Huawei / Cheng, Xueqi

2021

Abstract: Information seeking is an essential step for open-domain question answering to efficiently gather evidence from a large corpus. Recently, iterative approaches have been proven to be effective for complex questions, by recursively retrieving new evidence ... ...

Abstract	Information seeking is an essential step for open-domain question answering to efficiently gather evidence from a large corpus. Recently, iterative approaches have been proven to be effective for complex questions, by recursively retrieving new evidence at each step. However, almost all existing iterative approaches use predefined strategies, either applying the same retrieval function multiple times or fixing the order of different retrieval functions, which cannot fulfill the diverse requirements of various questions. In this paper, we propose a novel adaptive information-seeking strategy for open-domain question answering, namely AISO. Specifically, the whole retrieval and answer process is modeled as a partially observed Markov decision process, where three types of retrieval operations (e.g., BM25, DPR, and hyperlink) and one answer operation are defined as actions. According to the learned policy, AISO could adaptively select a proper retrieval action to seek the missing evidence at each step, based on the collected evidence and the reformulated query, or directly output the answer when the evidence set is sufficient for the question. Experiments on SQuAD Open and HotpotQA fullwiki, which serve as single-hop and multi-hop open-domain QA benchmarks, show that AISO outperforms all baseline methods with predefined strategies in terms of both retrieval and answer evaluations. Comment: Accepted at EMNLP 2021
Keywords	Computer Science - Computation and Language
Subject code	006
Publishing date	2021-09-14
Publishing country	us
Document type	Book ; Online
Database	BASE - Bielefeld Academic Search Engine (life sciences selection)

Full text online

Full text

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Article: Analysis of ANGPTL8 promoter activity and screening of related transcription factors in bovine

Wei, Xuefeng / Zhu, Yunchang / Du, Jie / Ma, Xiaojie / Zhao, Xue / Ma, Yaoyao / Han, Shuang / Ma, Yun

Gene. 2021 June 05, v. 784

2021

Abstract: Analysing the molecular regulation mechanism of fat deposition in yellow cattle can provide a theoretical basis for the breeding of excellent beef cattle. ANGPTL8 (angiopoietin-like protein 8) promotes the formation of lipid droplets during adipocyte ... ...

Abstract	Analysing the molecular regulation mechanism of fat deposition in yellow cattle can provide a theoretical basis for the breeding of excellent beef cattle. ANGPTL8 (angiopoietin-like protein 8) promotes the formation of lipid droplets during adipocyte differentiation. To explore the promoter active region of ANGPTL8 and predict potential transcription factors, we further provide a theoretical basis for the functional analysis and regulatory mechanism of ANGPTL8 in adipogenesis. The promoter region of bovine ANGPTL8 was cloned by overlap extension PCR. Online software was used to predict potential transcription factor binding sites, and it identified PPARγ, SREBP1, C/EBPα, and Znf423 transcription factor binding sites in ANGPTL8 promoter region. A luciferase reporter gene vector which contained different deletion fragments of the ANGPTL8 promoter was constructed. Then, the vectors were cotransfected into 293 T cells with the internal control plasmid pRL-TK by cationic liposomes, and the relative fluorescence intensity was detected by a microplate reader. The results of the luciferase activity analysis showed that the core promoter area of ANGPTL8 was in the −885/-227 bp region of the 5′ flanking sequence, while just two SREBP1 binding sites occurred in this area. When SREBP1 was knocked down by siRNA, the expression level of ANGPTL8 was reduced, and we speculated that SREBP1 may be an important transcription factor regulating ANGPTL8 transcription.
Keywords	adipocytes ; adipogenesis ; beef cattle ; computer software ; fluorescence ; lipids ; luciferase ; plasmids ; promoter regions ; reporter genes
Language	English
Dates of publication	2021-0605
Publishing place	Elsevier B.V.
Document type	Article
Note	NAL-AP-2-clean
ZDB-ID	391792-7
ISSN	1879-0038 ; 0378-1119
ISSN (online)	1879-0038
ISSN	0378-1119
DOI	10.1016/j.gene.2021.145594
Database	NAL-Catalogue (AGRICOLA)

Full text online

Accessible to users with ZB MED library card

In stock of ZB MED Bonn / Germany

Z 79/715: Show issues

Order via subito

Inter-library loan at ZB MED

Your chosen title can be delivered directly to ZB MED Cologne location if you are registered as a user at ZB MED Cologne.

Details ▾

Article: Buffalo long non-coding RNA gene11007 promotes myoblasts proliferation.

Zhang, Ning / Xu, Gaoxiao / Sun, Ping / Wang, Shuzhe / Zhu, Yunchang / Duan, Saixing / Jiang, Mingsheng / Li, Hui / Wei, Xuefeng / Ma, Yun

Frontiers in veterinary science

2022 Volume 9, Page(s) 857044

Abstract: Buffalo meat is of good quality because it is lean and tender, and could bring significant cardiovascular benefits. The underlying difference in muscle development and meat quality is a complex and precisely orchestrated process which has been ... ...

Abstract	Buffalo meat is of good quality because it is lean and tender, and could bring significant cardiovascular benefits. The underlying difference in muscle development and meat quality is a complex and precisely orchestrated process which has been demonstrated to be regulated by long non-coding RNAs (lncRNAs). However, the regulatory role of lncRNAs in the growth and development of buffalo skeletal muscle is still unclear. In this study, the Ribo-Zero RNA-Seq method was used to explore the lncRNA expression profiles of buffalo myoblasts during the proliferation and differentiation phases. A specific set of 9,978 lncRNAs was found. By comparing the expression profiles of lncRNAs, it was found that there were 1,576 differentially expressed lncRNAs (DELs) during buffalo myoblast differentiation. Twelve DELs were chosen and subsequently verified in eight different buffalo tissues during fetal and adult stages by using qPCR. Gene11007 was found to be one of the most down-regulated lncRNAs during buffalo myoblasts differentiation and it was subsequently characterized. EdU, CCK-8, qPCR and western blotting assays showed that gene11007 promoted the proliferation of buffalo myoblasts but it had no effect on cell differentiation. Our research may enrich the genome annotations of buffalo and provide a new molecular target for the in-depth understanding of the regulation of lncRNAs in skeletal muscle.
Language	English
Publishing date	2022-08-05
Publishing country	Switzerland
Document type	Journal Article
ZDB-ID	2834243-4
ISSN	2297-1769
ISSN	2297-1769
DOI	10.3389/fvets.2022.857044
Database	MEDical Literature Analysis and Retrieval System OnLINE

Order via subito

Article ; Online: MiR-107 Regulates Adipocyte Differentiation and Adipogenesis by Targeting Apolipoprotein C-2 (

Wei, Xuefeng / Zhao, Xue / Shan, Xinyue / Zhu, Yunchang / Wang, Shuzhe / Chen, Hong / Li, Hui / Ma, Yun

Genes

2022 Volume 13, Issue 8

Abstract	Adipogenesis is a complex and precisely orchestrated process mediated by a series of adipogenic regulatory factors. Recent studies have highlighted the importance of microRNAs (miRNAs) in diverse biological processes, most specifically in regulating cell differentiation and proliferation. However, the mechanisms of miRNAs in adipogenesis are largely unknown. In this study, we found that miR-107 expression was higher in bovine adipose tissue than that in other tissues, and there was a downregulation trend during adipocyte differentiation. To explore the function of miR-107 in adipocyte differentiation, agomiR-107 and antiagomiR-107 were transfected into bovine adipocytes, respectively. Oil Red O staining, CCK-8, EdU assays, RT-qPCR, and Western blotting were performed, and the results showed that overexpressed miR-107 significantly suppressed fat deposition and adipocyte differentiation, while knockdown of miR-107 promoted fat deposition and adipocytes differentiation. In addition, through bioinformatics analysis, luciferase reporter assays, RT-qPCR, and Western blotting, we identified apolipoprotein 2 (
MeSH term(s)	Adipocytes/metabolism ; Adipogenesis/genetics ; Animals ; Apolipoprotein C-II/metabolism ; Cattle ; Cell Differentiation/genetics ; MicroRNAs/genetics ; MicroRNAs/metabolism
Chemical Substances	Apolipoprotein C-II ; MicroRNAs
Language	English
Publishing date	2022-08-17
Publishing country	Switzerland
Document type	Journal Article ; Research Support, Non-U.S. Gov't
ZDB-ID	2527218-4
ISSN	2073-4425 ; 2073-4425
ISSN (online)	2073-4425
ISSN	2073-4425
DOI	10.3390/genes13081467
Database	MEDical Literature Analysis and Retrieval System OnLINE

Order via subito

Article ; Online: Transcriptome profiling of mRNAs in muscle tissue of Pinan cattle and Nanyang cattle.

Wei, Xuefeng / Zhu, Yunchang / Zhao, Xue / Zhao, Yadi / Jing, Yujia / Liu, Gege / Wang, Shuzhe / Li, Hui / Ma, Yun

Gene

2022 Volume 825, Page(s) 146435

Abstract	Mammalian muscle development is regulated by complex gene networks at the molecular level. The revelation of gene regulatory mechanisms is an important basis for the study of muscle development and molecular breeding. To analyze the excellent meat performance of Pinan cattle at the molecular level, we performed high-throughput RNA sequencing to analyze the key regulatory genes that determine the muscle quality traits in Pinan cattle (n = 3) and Nanyang cattle (n = 3). We identified 57 differentially expressed genes in muscle tissue of Pinan cattle compared to that of Nanyang cattle, including 32 upregulated and 25 downregulated genes. GO enrichment analysis showed that these genes were significantly enriched in 'molecular function', including voltage-gated ion channel activity, calcium channel activity and calcium ion binding, and KEGG pathway analysis results revealed that adrenergic signaling in cardio myocytes, cell adhesion molecules and inositol phosphate metabolism pathway were significantly enriched. We identified the reliability of RNA-Seq data through RT-qPCR. Meanwhile, we found that GSTA3, PLCB1 and ISYNA1 genes are highly expressed in muscle tissue of Pinan cattle, and these genes play important roles in PI3K/Akt, MEK1/2-ERK and p53-ISYNA1 signaling pathway. In summary, our results suggested that these differentially expressed genes may play important roles in muscle development in Pinan cattle. However, the functions and mechanism of these significantly differential expressed genes should be investigated in future studies.
MeSH term(s)	Animals ; Cattle/genetics ; Gene Expression Profiling/methods ; Mammals/genetics ; Muscles ; Phosphatidylinositol 3-Kinases/genetics ; RNA, Messenger/genetics ; Reproducibility of Results ; Transcriptome
Chemical Substances	RNA, Messenger
Language	English
Publishing date	2022-03-15
Publishing country	Netherlands
Document type	Journal Article
ZDB-ID	391792-7
ISSN	1879-0038 ; 0378-1119
ISSN (online)	1879-0038
ISSN	0378-1119
DOI	10.1016/j.gene.2022.146435
Database	MEDical Literature Analysis and Retrieval System OnLINE

Full text online

Accessible to users with ZB MED library card

In stock of ZB MED Cologne/Königswinter

Zs.A 1357: Show issues

Location:
Je nach Verfügbarkeit (siehe Angabe bei Bestand)
bis Jg. 1994: Bestellungen von Artikeln über das Online-Bestellformular
Jg. 1995 - 2021: Lesesall (1.OG)
ab Jg. 2022: Lesesaal (EG)

Order via subito

Details ▾

To top

Full text online

More links

Kategorien

Inter-library loan at ZB MED

Full text online

More links

Kategorien

Inter-library loan at ZB MED

More links

Kategorien

Order via subito

Full text online

More links

Kategorien

In stock of ZB MED Bonn / Germany

Order via subito

Inter-library loan at ZB MED

Full text online

More links

Kategorien

Inter-library loan at ZB MED

Full text online

More links

Kategorien

Inter-library loan at ZB MED

Full text online

More links

Kategorien

In stock of ZB MED Bonn / Germany

Order via subito

Inter-library loan at ZB MED

More links

Kategorien

Order via subito

More links

Kategorien

Order via subito

Full text online

More links

Kategorien

In stock of ZB MED Cologne/Königswinter

Order via subito