eprintid: 10132042 rev_number: 14 eprint_status: archive userid: 608 dir: disk0/10/13/20/42 datestamp: 2021-07-30 10:58:20 lastmod: 2021-12-24 23:22:00 status_changed: 2021-07-30 10:58:20 type: article metadata_visibility: show creators_name: Purver, M creators_name: Sadrzadeh, M creators_name: Kempson, R creators_name: Wijnholds, G creators_name: Hough, J title: Incremental Composition in Distributional Semantics ispublished: inpress divisions: UCL divisions: B04 divisions: C05 divisions: F48 keywords: Incrementality, Semantics, Vector space semantics, Incremental disambiguation note: Despite the incremental nature of Dynamic Syntax (DS), the semantic grounding of it remains that of predicate logic, itself grounded in set theory, so is poorly suited to expressing the rampantly context-relative nature of word meaning, and related phenomena such as incremental judgements of similarity needed for the modelling of disambiguation. Here, we show how DS can be assigned a compositional distributional semantics which enables such judgements and makes it possible to incrementally disambiguate language constructs using vector space semantics. Building on a proposal in our previous work, we implement and evaluate our model on real data, showing that it outperforms a commonly used additive baseline. In conclusion, we argue that these results set the ground for an account of the non-determinism of lexical content, in which the nature of word meaning is its dependence on surrounding context for its construal. abstract: Despite the incremental nature of Dynamic Syntax (DS), the semantic grounding of it remains that of predicate logic, itself grounded in set theory, so is poorly suited to expressing the rampantly context-relative nature of word meaning, and related phenomena such as incremental judgements of similarity needed for the modelling of disambiguation. Here, we show how DS can be assigned a compositional distributional semantics which enables such judgements and makes it possible to incrementally disambiguate language constructs using vector space semantics. Building on a proposal in our previous work, we implement and evaluate our model on real data, showing that it outperforms a commonly used additive baseline. In conclusion, we argue that these results set the ground for an account of the non-determinism of lexical content, in which the nature of word meaning is its dependence on surrounding context for its construal. date: 2021-07-07 date_type: published publisher: SPRINGER official_url: http://dx.doi.org/10.1007/s10849-021-09337-8 oa_status: green full_text_type: pub language: eng primo: open primo_central: open_green verified: verified_manual elements_id: 1878083 doi: 10.1007/s10849-021-09337-8 lyricists_name: Sadrzadeh, Mehrnoosh lyricists_id: MSADR73 actors_name: Flynn, Bernadette actors_id: BFFLY94 actors_role: owner full_text_status: public publication: Journal of Logic, Language and Information pages: 28 citation: Purver, M; Sadrzadeh, M; Kempson, R; Wijnholds, G; Hough, J; (2021) Incremental Composition in Distributional Semantics. Journal of Logic, Language and Information 10.1007/s10849-021-09337-8 <https://doi.org/10.1007/s10849-021-09337-8>. (In press). Green open access document_url: https://discovery.ucl.ac.uk/id/eprint/10132042/1/Purver2021_Article_IncrementalCompositionInDistri.pdf