0
Deep contextualized word representations
Abstract这篇文章是NAACL2018的Outstanding Paper。这篇文章的测试任务包括了情感分析。
标签：: Word Vector
0
Integrating Distributional Lexical Contrast into Word Embeddings for Antonym-Synonym Distinction
他们想增强一些能够明显区分同义词和反义词的特征。
0
Ontologically Grounded Multi-sense Representation Learning for Semantic Vector Space Models
IntroductionIn this paper thye propose two novel and general approaches for generating sense-specific word embeddings that are grounded in an ontology. 但是通用的词向量只为一个词分配一个词向量，无法解决解决一词多义的问题。 虽然Yu and Dre
0
AutoExtend Extending Word Embeddings to Embeddings for Synsets and Lexemes
Introduction这篇文章提出了一个叫做AutoExtand的方法，来运用其他的信息来增强word embedding的性能。用于学习用来表示synsets和lexemes。用于一些公开的知识库上，比如WordNet, Wikipedia和Freebase。 synset是指一个词语集合，这里面的词可以在一定的条件下相互替换。lexeme会将一个特定的拼写和发音和一个特定的意思匹配在一起。也
0
Semantic Specialisation of Distributional Word Vector Spaces using Monolingual and Cross-Lingual Constraints
这篇文章想用同一语言和跨语言资源中获得的约束来增强词向量的质量。
0
Information-Theory Interpretation of the Skip-Gram Negative-Sampling Objective Function
A dependency measure based on Jensen-ShannonThey define a dependency measure between two random variables, which is based on the Jensen-Shannon divergence. The Kullback-Leibler (KL) divergence of a d
0
Improved Word Representation Learning with Sememes
MethologyHowNetHowNet annotates precise senses to each word, and for each sense, HowNet annotates the significance of parts and attributes represented by sememes.
0
Not All Contexts Are Created Equal:Better Word Representations with Variable Attention
AbstractThey introduce an extension to the bag-of-words model for learning words represen- tations that take into account both syntactic and semantic properties within language. IntroductionFor BOW an
0
Modeling Word Meaning in Context with Substitute Vectors
AbstractContext representations are a key element in distributional models of word meaning. A recently proposed approach suggests to represent a context of a target word by a substitute vector. In thi
0
Adversarial Multi-task Learning for Text Classification
IntroductionMost existing work on multi-task learning attempts to divide the features of different tasks into private and shared spaces, merely based on whether parameters of some components should be