Graded Entailment for Compositional Distributional Semantics

Bankova, Desislava; Coecke, Bob; Lewis, Martha; Marsden, Daniel

doi:10.48550/arxiv.1601.04908

Cited by 16 publications

(26 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The data sets used in research on this topic tend to be either fully formal, focusing on logic instead of natural language Allamanis et al, 2016), or fully natural, as is the case for manually annotated data sets of English sentence pairs such as SICK (Marelli et al, 2014) or SNLI (Bowman et al, 2015a). Moreover, entailment recognition models are often endowed with functionality reflecting pre-established linguistic or semantic regularities of the data (Bankova et al, 2016;Serafini and Garcez, 2016;Sadrzadeh et al, 2018). Recently, Shen et al (2018) showed that recurrent networks can learn to recognize logical inference relations if they are extended with a bias towards modelling hierarchical structures.…”

Section: Introduction and Related Workmentioning

confidence: 99%

Siamese recurrent networks learn first-order logic reasoning and exhibit zero-shot compositional generalization

Mul,

Zuidema

2019

Preprint

View full text Add to dashboard Cite

Can neural nets learn logic? We approach this classic question with current methods, and demonstrate that recurrent neural networks can learn to recognize first order logical entailment relations between expressions. We define an artificial language in first-order predicate logic, generate a large dataset of sample 'sentences', and use an automatic theorem prover to infer the relation between random pairs of such sentences. We describe a Siamese neural architecture trained to predict the logical relation, and experiment with recurrent and recursive networks. Siamese Recurrent Networks are surprisingly successful at the entailment recognition task, reaching near perfect performance on novel sentences (consisting of known words), and even outperforming recursive networks. We report a series of experiments to test the ability of the models to perform compositional generalization. In particular, we study how they deal with sentences of unseen length, and sentences containing unseen words. We show that set-ups using LSTMs and GRUs obtain high scores on these tests, demonstrating a form of compositionality.Preprint. Under review.

show abstract

Section: Introduction and Related Workmentioning

confidence: 99%

Siamese recurrent networks learn first-order logic reasoning and exhibit zero-shot compositional generalization

Mul,

Zuidema

2019

Preprint

View full text Add to dashboard Cite

show abstract

“…In addition, our DisCoCat-based QNLP framework is naturally generalisable to accommodate mapping sentences to quantum circuits involving mixed states and quantum channels. This is useful as mixed states allow for modelling lexical entailement and ambiguity [55,56].…”

Section: To Clearly Show the Decrease In Training And Test Errors As ...mentioning

confidence: 99%

Grammar-aware sentence classification on quantum computers

Meichanetzidis¹,

Toumi²,

Felice³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

“…It has been extensively studied, both axiomatically [6,7,9,14] and concretely [10,15,16,17]. Recently, applications of the CPM construction in the context of compositional distributional models of meaning [4,5,15,17] have prompted renewed interest on iterated CPM constructions [3], with the discovery of new features due to their additional degrees of freedom [19].…”

Section: Introductionmentioning

confidence: 99%

Higher-order CPM Constructions

Gogioso

2019

Electron. Proc. Theor. Comput. Sci.

View full text Add to dashboard Cite

We define a higher-order generalisation of the CPM construction based on arbitrary finite abelian group symmetries of symmetric monoidal categories. We show that our new construction is functorial, and that its closure under iteration can be characterised by seeing the construction as an algebra for an appropriate monad. We provide several examples of the construction, connecting to previous work on the CPM construction and on categorical probabilistic theories, as well as upcoming work on higher-order interference and hyper-decoherence. *

show abstract

Graded Entailment for Compositional Distributional Semantics

Cited by 16 publications

References 0 publications

Siamese recurrent networks learn first-order logic reasoning and exhibit zero-shot compositional generalization

Siamese recurrent networks learn first-order logic reasoning and exhibit zero-shot compositional generalization

Grammar-aware sentence classification on quantum computers

Higher-order CPM Constructions

Contact Info

Product

Resources

About