Do Multi-Sense Embeddings Improve Natural Language Understanding?
Explore this paper's citation graph
Summary
A multisense embedding model based on Chinese Restaurant Processes is introduced that achieves state of the art performance on matching human word similarity judgments, and a pipelined architecture for incorporating multi-sense embeddings into language understanding is proposed.
- Type
- preprint
- Published
- 2015-06-02
- Cited by
- 235
- References
- 38
- Access
- Open access
- OpenAlex
- https://openalex.org/W1010415138
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:6222768
Keywords
Computer science, Sense (electronics), Natural language, Natural (archaeology), Natural language processing
References
- Recurrent neural network based language model
- Efficient Estimation of Word Representations in Vector Space
- When Are Tree Structures Necessary for Deep Learning of Representations?
- Semantic Compositionality through Recursive Matrix-Vector Spaces
- Learning State Space Trajectories in Recurrent Neural Networks
- Exchangeable and partially exchangeable random partitions
- Placing search in context: the concept revisited
- Long Short-Term Memory
- A Bayesian Analysis of Some Nonparametric Problems
- Contextual correlates of synonymy
- Three new graphical models for statistical language modelling
- SemEval-2010 Task 8: Multi-Way Classification of Semantic Relations Between Pairs of Nominals
- Improved Semantic Representations From Tree-Structured Long Short-Term Memory Networks
- A unified architecture for natural language processing: deep neural networks with multitask learning
- Hierarchical Topic Models and the Nested Chinese Restaurant Process
- Deep Recursive Neural Networks for Compositionality in Language
- Hierarchical Dirichlet Processes
- Efficient Non-parametric Estimation of Multiple Embeddings per Word in Vector Space
- Improving Word Representations via Global Context and Multiple Word Prototypes
- Multi-Prototype Vector-Space Models of Word Meaning
Cited by
- Contextual Text Understanding in Distributional Semantic Space
- A Joint Segmentation and Classification Framework for Sentence Level Sentiment Classification
- Deep learning for sentiment analysis: successful approaches and future challenges
- Sentiment Embeddings with Applications to Sentiment Analysis
- Bilingual Learning of Multi-sense Embeddings with Discrete Autoencoders
- Multi-phase Word Sense Embedding Retrofitting with Lexical Ontology
- Learning Word Sense Embeddings from Word Sense Definitions
- Intrinsic Subspace Evaluation of Word Embedding Representations
- Making Sense of Word Embeddings
- Supersense Embeddings: A Unified Model for Supersense Interpretation, Prediction, and Utilization
- Evaluating multi-sense embeddings for semantic resolution monolingually and in word translation
- Nasari: Integrating explicit knowledge and corpus statistics for a multilingual representation of concepts and entities
- Multi-phase Word Sense Embedding Learning Using a Corpus and a Lexical Ontology
- Geometry of Polysemy
- Topic-Aware Deep Compositional Models for Sentence Classification
- Empirical studies on word representations
- Context-Dependent Sense Embedding
- Embedding Words and Senses Together via Joint Knowledge-Enhanced Training
- Real Multi-Sense or Pseudo Multi-Sense: An Approach to Improve Word Representation
- Can Topic Modelling benefit from Word Sense Information?