Solving Arithmetic Word Problems of Entailing Deep Implicit Relations by Qualia Syntax-Semantic Model

Solving arithmetic word problems that entail deep implicit relations is still a challenging problem.However,significant progress has been made in solving Arithmetic Word Problems(AWP)over the past six decades.This paper proposes to discover deep implicit relations by qualia inference to solve Arithmetic Word Problems entailing Deep Implicit Relations(DIR-AWP),such as entailing commonsense or subject-domain knowledge involved in the problem-solving process.This paper proposes to take three steps to solve DIR-AWPs,in which the first three steps are used to conduct the qualia inference process.The first step uses the prepared set of qualia-quantity models to identify qualia scenes from the explicit relations extracted by the Syntax-Semantic(S2)method from the given problem.The second step adds missing entities and deep implicit relations in order using the identified qualia scenes and the qualia-quantity models,respectively.The third step distills the relations for solving the given problem by pruning the spare branches of the qualia dependency graph of all the acquired relations.The research contributes to the field by presenting a comprehensive approach combining explicit and implicit knowledge to enhance reasoning abilities.The experimental results on Math23K demonstrate hat the proposed algorithm is superior to the baseline algorithms in solving AWPs requiring deep implicit relations.

An Article Review： Impact of Explicit Instruction on EFL Learners＇ Implicit and Explicit Knowledge： A Case of English Relative Clauses

This review discusses Nezakat-Alhossaini, Youhanaee, and Moinzadeh＇s research study entitled ＂Impact of Explicit Instruction on EFL Learners＇ Implicit and Explicit Knowledge： A Case of English Relative Clauses.＂ This study was chosen for evaluation because it strives to attach significance to explicit instruction in L2 acquisition, unlike other more recent research, which seeks to reinforce implicit instruction as it is viewed as the idealistic goal of language learning （Rebuschat ＆ William, 2009）. The present review will be developed by means of an evaluation of Alhossaini and her colleagues＇ study, consisting of a concise summary of the study, a classification of the philosophical perspective, the selection of criteria, and the strengths and weaknesses of the study. In this review, I hope that I succeed to broadly navigate the research enterprise, commencing with the philosophical perspective of research, such as the epistemological and ontological stances shaping the philosophical perspective and then colouring the research. By reviewing this study, I would also hope that I successfully evaluate the research quality by using appropriate criteria in an attempt to suggest potential directions for further research （under strengths and weaknesses of the study）.

FIXED POINT THEOREMS FOR MAPPINGS SATISFYING IMPLICIT RELATION ON TWO COMPLETE AND COMPACT METRIC SPACES

First, the implicit relations were given. A common fixed point theorem was proved for two mappings satisfying implicit relation functions. A further fixed point theorem was proved for mappings satisfying implicit relation functions on two compact metric spaces.

Common New Fixed Point Theorem in Modified Intuitionistic Fuzzy Metric Spaces Using Implicit Relation

In this paper, we prove some common fixed point theorems for two pair of compatible and subsequentially continuous mappings satisfying an implicit relation in Modified Intuitionistic fuzzy metric spaces. Consequently, our results improve and sharpen many known common fixed point theorems available in the existing literature of metric fixed point theory.

Fixed Point Theorems in Intuitionistics Fuzzy Metric Spaces Using Implicit Relations

In this paper, we proved some fixed point theorems in intuitionistic fuzzy metric spaces applying the properties of weakly compatible mapping and satisfying the concept of implicit relations for t norms and t connorms.

融合特征编码和短语交互感知的隐式篇章关系识别

隐式篇章关系识别难度大、普遍性高.从论元编码和论元交互角度入手,提出了一种融合特征编码和短语交互感知的隐式篇章关系识别模型.该模型兼顾了论元本身特征和论元间交互特征的作用,并分别进行了优化.论元编码部分整合了双向长短时记忆网络和循环注意力卷积神经网络,能够更全面地捕获论元全局和局部特征;论元交互部分从短语层级考虑论元间的语义关系建模,构建了短语级交互注意力机制,并利用神经张量网络深入挖掘其中的关系模式,更能体现出论元间潜在的更深层次的关联关系.在宾州篇章树库数据集上的实验结果表明,该模型F1值均优于其他模型.

融合词语语义与标签依赖的隐式篇章关系识别

中文隐式篇章关系识别旨在推断出两个论元间的篇章关系类型。然而,现有的方法往往忽略了论元中词语所蕴含的关键信息,并且仅考虑单个层级内的篇章关系类型,忽略了各层级间篇章关系的依赖关联。鉴于此,提出融合词语语义和标签依赖的方法,以序列生成的方式实现篇章关系识别,先根据相似度权重将词向量嵌入到字编码表示中,应用字词对齐注意力机制强调关键字、词信息,再采用标签注意力编码从蕴含词语语义的论元表示和篇章关系表示中获取篇章关系依赖性的上下文表示,以自下而上的方式预测顶层的篇章关系类型。此外,构建面向阅读理解篇章的篇章关系数据集,并在该数据集上展开实验,结果显示隐式篇章关系识别准确率和F1值分别达到74.19%和73.81%,最终验证了该方法的有效性。

基于多视角对比学习的隐式篇章关系识别

隐式篇章关系识别IDRR的相关工作集中在篇章单元编码器的设计上。将对比学习引入到IDRR,以此获得区分度更高的篇章单元表征。具体地,首先使用一个轻量的IDRR模型;然后为了学习到更好的篇章单元表征,分别从样例层级、批层级和群层级,探索了3种不同视角的对比学习方法在IDRR中的应用;最后本文将多视角对比学习目标联合IDRR同时进行训练。本文提出的方法几乎不增加训练时间,而且只引入少量额外参数。基于PDTB 2.0的实验结果表明该方法达到了最优性能。

Discriminative explicit instance selection for implicit discourse relation classification

Discourse relation classification is a fundamental task for discourse analysis,which is essential for understanding the structure and connection of texts.Implicit discourse relation classification aims to determine the relationship between adjacent sentences and is very challenging because it lacks explicit discourse connectives as linguistic cues and sufficient annotated training data.In this paper,we propose a discriminative instance selection method to construct synthetic implicit discourse relation data from easy-to-collect explicit discourse relations.An expanded instance consists of an argument pair and its sense label.We introduce the argument pair type classification task,which aims to distinguish between implicit and explicit argument pairs and select the explicit argument pairs that are most similar to natural implicit argument pairs for data expansion.We also propose a simple label-smoothing technique to assign robust sense labels for the selected argument pairs.We evaluate our method on PDTB 2.0 and PDTB 3.0.The results show that our method can consistently improve the performance of the baseline model,and achieve competitive results with the state-of-the-art models.

面向中医古籍的隐式关系抽取方法研究

自然语言种类丰富、形式灵活多变的特征使得隐式关系抽取成为目前关系抽取领域中富有难度和挑战性的任务之一。通过引入构式语法理论和依存句法分析两种认知语言学范畴的理论技术,构建了一种面向中医古籍中隐式关系的抽取方法。首先利用构式语法理论制定文本构式化策略、分析并定义出8种构式特征与5种构式类型,并使用CART(classification and regression tree,CART)分类模型完成文本分类;其次对其中4类构式使用依存句法分析技术构建句法树,通过分析句法树中的特定结构,制定医学类实体间的关系三元组抽取规则,实现隐式关系抽取;最后在经典中医古籍《黄帝内经》数据集上进行测试,实验结果表明了方法的有效性。

让步构式“X归X”的多维考察

用于让步-转折关系复句的构式“X归X”,表示对事实的让步性认同,以便引出后续句S。后续句S表示的意思往往跟利用语境信息从“X归X”中X激活的信息推出的最可能的结论相反。整个构式有隐性否定作用,强调异态共存。该构式中的X虽然通常是动词、形容词等谓词性成分,但也可以是能激活事件解读的名词性成分。其中的“X归X”也可以是两项或多项平列,但是多个X之间存在同义或类义关系,而不可能是反义关系。整个让转复句“X归X,S”经常用在观点博弈或立场对立语境中。

VQT:value cardinality and query pattern based R-schema to XML schema translation with implicit referential integrity

In this paper,we propose a new relational schema (R-schema) to XML schema translation algorithm, VQT, which analyzes the value cardinality and user query patterns and extracts the implicit referential integrities by using the cardinality property of foreign key constraints between columns and the equi-join characteristic in user queries. The VQT algorithm can apply the extracted implied referential integrity relation information to the R-schema and create an XML schema as the final result. Therefore, the VQT algorithm prevents the R-schema from being incorrectly converted into the XML schema, and it richly and powerfully represents all the information in the R-schema by creating an XML schema as the translation result on behalf of the XML DTD.

Set-Valued Non-Linear Random Implicit Quasivariational Inclusions

In this paper, we propose iterative algorithms for set valued nonlinear random implicit quasivariational inclusions. We define the related random implicit proximal operator equations and establish an equivalence between them. Finally, we prove the existence and convergence of random iterative sequences generated by random iterative algorithms.

汉语“N的V”结构加工的韵律制约

已有研究表明内隐韵律能够促进书面文本的理解,但以往研究多集中在重音节拍语言,而较少关注汉语这类音节节拍语言。为了探究内隐韵律作用的普遍性,研究采用眼动技术探讨了句子阅读中“N的V”结构加工的韵律制约。研究发现,“N的V”结构加工中存在稳定的韵律形式效应:动词为单音节时比动词为双音节时的注视时间更长,回视更多。韵律形式和语义关系之间存在交互作用,表现在受事语义关系条件下,在反映早期加工的凝视时间上,单音节动词条件长于双音节动词条件。但这种交互作用在晚期加工阶段则不受语义关系影响。表明“N的V”结构加工中的韵律形式效应稳定且持续。研究支持内隐韵律假说。

面向社交推荐的自适应高阶隐式关系建模

近年来,社交推荐的研究主要聚焦于社交网络中显式、隐式关系的联合建模,却忽视了高阶隐式关系并非对每个用户都同等重要这一特殊现象.高阶隐式关系对一个有着足够多邻居的用户与一个仅有少量邻居的用户重要性存在明显差异.此外,由于社交关系建立的随机性,显式关系并不总是可用的.提出了一种新的自适应高阶隐式关系建模方法(adaptive high-order implicit relations modeling,AHIRM),该模型由3个部分组成:首先,过滤不可靠关系且识别出潜在可靠关系.旨在避免不可靠关系带来的负面影响,并部分缓解数据稀疏的问题;其次,设计自适应随机游走算法,结合规范化后的节点中心度为用户捕获不同阶数的邻居,构建用户间的高阶隐式关系,进而重构社交网络;最后,运用图卷积网络(graph convolutional network,GCN)聚合邻居节点信息,更新用户嵌入,实现高阶隐式关系建模,从而进一步缓解数据稀疏问题.在建模过程中,同时考虑到社交结构和个人偏好的影响,模拟并保留了社交影响传播的过程.在LastFM、Douban和Gowalla这3个数据集上与相关算法做了对比验证,结果证实了该模型的有效性和合理性.

基于语义图谱的图书馆文献推荐系统设计

针对图书馆文献推荐系统推荐方案效果不够理想的问题,该研究使用隐语义模型(Latent Factor Model,简称LFM)构建语义图谱的方式搭建混合隐语义推荐系统。选取数据集CiteULike来验证其应用效果。结果显示,在推荐文献数量为20个时,基于混合LFM、朴素LFM、协同过滤、ResNet卷积神经网络的各推荐系统平均召回率、平均准确率分别为63.2%、47.6%、59.2%、56.5%和61.6%、43.9%、58.8%、58.9%,且基于混合LFM的推荐系统计算耗时146.2s相对较小,相对NDCG值0.635为各方案最高。数据表明,该推荐系统在推荐文献数量较小时,实用性更高,计算耗时较少。

大学生对不同社会关系的内隐合作态度及属性认知

目的探索大学生对不同社会关系的内隐合作态度。方法选取70名在校本科生参加单靶内隐联想测验(ST-IAT)和Simon任务。结果①从内隐层面,大学生更愿意和有社会关系的人合作,并在合作指数上表现出差别有序性,从高到低分别为亲缘关系>地缘关系>学缘关系>陌生关系;②大学生更多地将亲缘关系和学缘关系与情感性相联结,而对地缘关系和陌生关系未发现显著的情感或工具倾向性。结论大学生对社会关系的合作偏好性更多地是建立在关系的情感性基础之上,社会互动可以改变关系的属性。

内隐自尊的ERP研究

研究使用GNAT范式对内隐自尊的ERP进行研究。结果发现:在自我-积极与自我-消极的感受性指标(d’)和反应时上存在显著差异;他人-积极与他人-消极的差异仅在d’上;自我-积极与自我-消极的P3在潜伏期和振幅上存在显著差异;他人-积极与他人-消极的P3之间的差异仅表现在潜伏期上。

内隐攻击的ERP研究

阈下启动是研究脑机制的重要手段之一。研究采用阈下启动刺激范式,选取华东师范大学本科生22人,分别记录64个电极点脑电数据。结果表明,阈下刺激的情况下N400;Fz的波幅最大,显著高于Cz的波幅和Pz的波幅。脑区和刺激图片之间的交互作用显著,脑区与刺激来源的交互作用显著;P300阈上刺激和阈下刺激的波幅上不存在显著差异,说明阈下刺激对P300的影响较小。

汉语隐式篇章关系识别

采用一个自建的汉语篇章结构语料库(隐式关系占80%)进行隐式关系识别。语料中将篇章关系分成3个层次,第一层包含因果、并列、转折、解说四大类。在此语料上,利用上下文特征、词汇特征、依存树特征,采用最大熵的分类方法对四大类关系进行识别。实验结果显示,总正确率为62.15%,其中并列类识别效果最好,F1值达到75.26%。