A practical and linguistically-motivated approach to compositional distributional semantics

Paperno, Denis; Baroni, Marco

doi:10.3115/v1/p14-1009

Cited by 54 publications

(84 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The issue of semantic compositionality is also actively discussed in the distributional semantics literature (see, e.g., (Mitchell and Lapata, 2010;Erk, 2013;Lewis and Steedman, 2013;Paperno et al, 2014)), investigating how to combine vectors. This could be seen as composition on the level of intensions (if one sees distributional representations as intensions, as is variously hinted at, e.g.…”

Section: Related Workmentioning

confidence: 99%

Simple Learning and Compositional Application of Perceptually Grounded Word Meanings for Incremental Reference Resolution

Kennington¹,

Schlangen²

2015

Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Confere

View full text Add to dashboard Cite

An elementary way of using language is to refer to objects. Often, these objects are physically present in the shared environment and reference is done via mention of perceivable properties of the objects. This is a type of language use that is modelled well neither by logical semantics nor by distributional semantics, the former focusing on inferential relations between expressed propositions, the latter on similarity relations between words or phrases. We present an account of word and phrase meaning that is perceptually grounded, trainable, compositional, and 'dialogueplausible' in that it computes meanings word-by-word. We show that the approach performs well (with an accuracy of 65% on a 1-out-of-32 reference resolution task) on direct descriptions and target/landmark descriptions, even when trained with less than 800 training examples and automatically transcribed utterances.

show abstract

Section: Related Workmentioning

confidence: 99%

Simple Learning and Compositional Application of Perceptually Grounded Word Meanings for Incremental Reference Resolution

Kennington¹,

Schlangen²

2015

Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Confere

View full text Add to dashboard Cite

show abstract

“…More recently, syntactic dependency-based compositional models have been proposed (Paperno et al, 2014;Socher et al, 2014;Tsubaki et al, 2013). One of the advantages of these models is that they are less restricted by word order.…”

Section: Related Workmentioning

confidence: 99%

Jointly Learning Word Representations and Composition Functions Using Predicate-Argument Structures

Hashimoto

Stenetorp

Misawa

et al. 2014

Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP)

View full text Add to dashboard Cite

We introduce a novel compositional language model that works on PredicateArgument Structures (PASs). Our model jointly learns word representations and their composition functions using bagof-words and dependency-based contexts. Unlike previous word-sequencebased models, our PAS-based model composes arguments into predicates by using the category information from the PAS. This enables our model to capture longrange dependencies between words and to better handle constructs such as verbobject and subject-verb-object relations. We verify this experimentally using two phrase similarity datasets and achieve results comparable to or higher than the previous best results. Our system achieves these results without the need for pretrained word vectors and using a much smaller training corpus; despite this, for the subject-verb-object dataset our model improves upon the state of the art by as much as ∼10% in relative performance.

show abstract

“…In future work we aim to extend our evaluation to more complex compositional distributional semantic models such as the lexical function model (Paperno et al, 2014) or the Anchored Packed Dependency Tree framework (Weir et al, 2016). We furthermore want to investigate how far the sense-discriminating ability of composition can be leveraged for other tasks.…”

Section: Resultsmentioning

confidence: 99%

Proceedings of the 1st Workshop on Sense, Concept and Entity Representations and their Applications

2017

View full text Add to dashboard Cite

PrefaceWelcome to the 1st Workshop on Sense, Concept and Entity Representations and their Applications (SENSE 2017). The aim of SENSE 2017 is to focus on addressing one of the most important limitations of word-based techniques in that they conflate different meanings of a word into a single representation. SENSE 2017 brings together researchers in lexical semantics, and NLP in general, to investigate and propose sense-based techniques as well as to discuss effective ways of integrating sense, concept and entity representations into downstream applications.The workshop is targeted at covering the following topics:• Utilizing sense/concept/entity representations in applications such as Machine Translation, Information Extraction or Retrieval, Word Sense Disambiguation, Entity Linking, Text Classification, Semantic Parsing, Knowledge Base Construction or Completion, etc.• Exploration of the advantages/disadvantages of using sense representations over word representations.• Proposing new evaluation benchmarks or comparison studies for sense vector representations.• Development of new sense representation techniques (unsupervised, knowledge-based or hybrid).• Compositionality of senses: learning representations for phrases and sentences.• Construction and use of sense representations for languages other than English as well as multilingual representations.We received 21 submissions, accepting 15 of them (acceptance rate: 71%).We would like to thank the Program Committee members who reviewed the papers and helped to improve the overall quality of the workshop. We also thank Aylien for their support in funding the best paper award. Last, a word of thanks also goes to our invited speakers, Roberto Navigli (Sapienza University of Rome) and Hinrich Schütze (University of Munich).Jose Camacho-Collados and Mohammad Taher Pilehvar Co-Organizers of SENSE 2017iii AbstractThis article describes a method to build semantic representations of composite expressions in a compositional way by using WordNet relations to represent the meaning of words. The meaning of a target word is modelled as a vector in which its semantically related words are assigned weights according to both the type of the relationship and the distance to the target word. Word vectors are compositionally combined by syntactic dependencies. Each syntactic dependency triggers two complementary compositional functions: the named head function and dependent function. The experiments show that the proposed compositional method performs as the state-of-the-art for subjectverb expressions, and clearly outperforms the best system for transitive subject-verbobject constructions. IntroductionThe principle of compositionality (Partee, 1984) states that the meaning of a complex expression is a function of the meaning of its constituent parts and of the mode of their combination. In the recent years, different distributional semantic models endowed with a compositional component have been proposed. Most of them define words as high-dimensional vectors where dimensions ...

show abstract

A practical and linguistically-motivated approach to compositional distributional semantics

Cited by 54 publications

References 8 publications

Simple Learning and Compositional Application of Perceptually Grounded Word Meanings for Incremental Reference Resolution

Simple Learning and Compositional Application of Perceptually Grounded Word Meanings for Incremental Reference Resolution

Jointly Learning Word Representations and Composition Functions Using Predicate-Argument Structures

Proceedings of the 1st Workshop on Sense, Concept and Entity Representations and their Applications

Contact Info

Product

Resources

About