Vector-based Models of Semantic Composition

By Jeff Mitchell, Mirella Lapata. 2008

806 citations recorded by the source corpus.

Subfields: Natural Language Processing, Representation & Self-Supervised Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

In Representation & Self-Supervised Learning

Citations in the corpus

Learning Distributed Representations of Sentences from Unlabelled Data (2016, background): “See the contrasting conclusions in (Mitchell and Lapata, 2008; Clark and Pulman, 2007; Baroni et al., 2014a; Milajevs et al., 2014) among others.”
Learning General Purpose Distributed Sentence Representations via Large Scale Multi-task Learning (2018, background): “For example, Mitchell & Lapata (2008) present an additive and multiplicative linear composition function of the distributed representations of individual words.”
A Convolutional Neural Network for Modelling Sentences (2014, background): “In some cases, composition is defined by algebraic operations over word meaning vectors to produce sentence meaning vectors (Erk and Padó, 2008; Mitchell and Lapata, 2008; Mitchell and Lapata, 2010; Turney, 2012; Erk, 2012; Clarke, 2012).”
Generating Informative and Diverse Conversational Responses via Adversarial Information Maximization (2018, background): “Similarly, the Average metric [42] considers the average embedding cosine similarity.”
A SICK cure for the evaluation of compositional distributional semantic models (2014, methodology): “Existing data sets, such as those introduced by Mitchell and Lapata (2008) and Grefenstette and Sadrzadeh (2011), are limited to a few hundred instances of very short sentences with a ﬁxed structure.”
Towards Universal Paraphrastic Sentence Embeddings (2015, background): “Mitchell & Lapata (2008; 2010) considered bigram compositionality, comparing many functions for composing two word vectors into a single vector to represent their bigram.”
A Hierarchical Latent Variable Encoder-Decoder Model for Generating Dialogues (2016, background): “age), Embedding Extrema (Extrema) and Embedding Greedy (Greedy) (Mitchell and Lapata 2008; Forgues et al. 2014; Rus and Lintean 2012).”
From Frequency to Meaning: Vector Space Models of Semantics (2010, background): “Mitchell and Lapata (2008) propose composition models sensitive to word order.”
Deep Unordered Composition Rivals Syntactic Methods for Text Classification (2015, methodology): “Mitchell and Lapata (2008) experiment with many of them to model the compositionality of short phrases.”
Grounding Action Descriptions in Videos (2013, methodology): “We compute inter-annotator agreement (and the forthcoming evaluation scores) using Spearman’s rank correlation coefficient (ρ), a non-parametric test which is widely used for similar evaluation tasks (Mitchell and Lapata, 2008; Bruni et al., 2011; Erk and McCarthy, 2009).”
PAWS: Paraphrase Adversaries from Word Scrambling (2019, background): “(1)-(5) are drawn from PAWS, and (6) is from Mitchell and Lapata (2008).”
A semantic matching energy function for learning with multi-relational data (2013, methodology): “More precisely, it is connected to models aiming at composing such vector embeddings for obtaining phrase or context representations (Mitchell and Lapata 2008), via tensor products (Smolensky 1990) or matrix operations (Paccanaro and Hinton 2001; Socher et al.”
Approximating Interactive Human Evaluation with Self-Play for Open-Domain Dialog Systems (2019, methodology): “Automatic metrics Embedding-based metrics compare generated sentences to ground truth sentences using a vector representation of words [2].”
Experimental Support for a Categorical Compositional Distributional Model of Meaning (2011, methodology): “Model Parameters As in the first experiment, the lexical vectors from (Mitchell and Lapata, 2008) were used for the other models evaluated (additive, multiplicative and baseline)8 and for the noun vec-”