Semantic Compositionality through Recursive Matrix-Vector Spaces

By Richard Socher, Brody Huval, Christopher D. Manning, Andrew Y. Ng. 2012

1,414 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

Semi-supervised Sequence Learning (2015, neutral): “MV-RNN [28] 21 .”
Distributed Representations of Words and Phrases and their Compositionality (2013, background): “Our work can thus be s e n as complementary to the existing approach that attempts to represent phrases using recursiv e matrix-vector operations [16].”
Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank (2013, methodology): “For the tree with (vector,matrix) nodes: (p 2 ,P 2 ) (a,A) (p 1 ,P 1 ) (b,B) (c,C) the MV-RNN computes the ﬁrst parent vector and its matrix via two equations: where W M ∈ R d × 2 d and the result is again a d × d matrix.”
Experience Grounds Language (2020, methodology): “A.M. Turing (Computing Machinery and Intelligence 1950) In the language community, this paradigm has been described under the “Speaker-Listener” model (Stephens et al., 2010), and a rich theory to describe this computationally is being actively developed under the Rational Speech Act Model (Frank…”
Revisiting LSTM Networks for Semi-Supervised Text Classification via Mixed Objective Function (2019, methodology): “Current state-of-the-art approaches for text classification involve using pretrained LSTMs (DL15; Dai and Le 2015) or complex computationally intensive models (JZ17; Johnson and Zhang 2017).”
Neural Turing Machines (2014, background): “Considering this property, we do not feel that it is urgent or even necessarily valuable to build explicit parse trees to merge composite structures greedily (Pollack, 1990) (Socher et al., 2012) (Frasconi et al.”
Grounded Compositional Semantics for Finding and Describing Images with Sentences (2014, background): “Since words rarely appear in isolation, the task of learning compositional meaning representations for longer phrases has recently received a lot of attention (Mitchell and Lapata, 2010; Socher et al., 2010; Socher et al., 2012; Grefenstette et al., 2013).”
Programmable Agents (2017, background): “Compositionality has become a topic of great interest in machine learning, robotics and cognitive science [20, 30, 19, 9, 35, 26], and it has been central to research in language [32, 25, 33, 36].”
Recurrent Continuous Translation Models (2013, background): “…sentiment, paraphrase or dialogue labels, signiﬁcantly beyond the word level and to accurately predict labels for a highly diverse range of unseen phrases and sentences (Grefenstette et al., 2011; Socher et al., 2011; Socher et al., 2012; Hermann and Blunsom, 2013; Kalchbrenner and Blunsom, 2013).”
Convolutional Neural Networks for Sentence Classification (2014, methodology): “MV-RNN : Matrix-Vector Recursive Neural Network with parse trees (Socher et al., 2012).”
A SICK cure for the evaluation of compositional distributional semantic models (2014, background): “…DSMs, or CDSMs) have been proposed, with the purpose of representing the meaning of phrases and sentences by composing the distributional representations of the words they contain (Mitchell and Lapata, 2010; Baroni and Zamparelli, 2010; Grefenstette and Sadrzadeh, 2011; Socher et al., 2012).”
Improved Semantic Representations From Tree-Structured Long Short-Term Memory Networks (2015, methodology): “, 2011), compose phrase representations from word vectors (Socher et al., 2012), and classify the sentiment polarity of sentences (Socher et al.”
Towards Universal Paraphrastic Sentence Embeddings (2015, background): “…averaging networks (DANs) (Iyyer et al., 2015), feature-weighted averaging (Yu & Dredze, 2015), recursive neural networks based on parse structure (Socher et al., 2011; 2012; 2013; ˙Irsoy & Cardie, 2014; Wieting et al., 2015), recursive networks based on non-syntactic hierarchical structure (Zhao…”
A Joint Many-Task Model: Growing a Neural Network for Multiple NLP Tasks (2016, background): “, 2016), relation classification (Socher et al., 2012), sentiment analysis (Socher et al.”
SemEval-2014 Task 1: Evaluation of Compositional Distributional Semantic Models on Full Sentences through Semantic Relatedness and Textual Entailment (2014, background): “…of DSMs (CDSMs) have been proposed, with the purpose of representing the meaning of phrases and sentences by composing the distributional representations of the words they contain (Baroni and Zampar-elli, 2010; Grefenstette and Sadrzadeh, 2011; Mitchell and Lapata, 2010; Socher et al., 2012).”
Self-Adaptive Hierarchical Sentence Model (2015, background): “Recursive autoencoder [Socher et al., 2011] and Matrix-vector recursive neural network [Socher et al., 2012].”
Illinois-LH: A Denotational and Distributional Approach to Semantics (2014, background): “…work that extends distributional similarities to phrases and sentences (Mitchell and Lapata, 2010; Baroni and Zamparelli, 2010; Grefenstette and Sadrzadeh, 2011; Socher et al., 2012), we deﬁne a phrase vector p to be the pointwise multiplication product of the vectors of the words in the phrase:”
Towards Linguistically Generalizable NLP Systems: A Workshop and Shared Task (2017, methodology): “This model, rather than parameter-izing the composition function by the words being composed (Socher et al., 2012), uses a single more powerful tensor-based composition function for composing each node of the syntactic tree.”