Limbäck-Stokin, Tilen Gaetano; Lo, Kin Ian; Sadrzadeh, Mehrnoosh; (2025) Degrees of Entanglement and Contextuality in Quantum Natural Language Processing. UCL Department of Computer Science: London, UK.

Preview

Text Quantum_Phenomena_NLP.pdf - Published Version Download (2MB) | Preview

Abstract

This paper regards an interdisciplinary area of study between physics and linguistics, and uses a setting which has been dubbed Quantum Natural Language Processing (QNLP). So far in QNLP, the focus has been on variational quantum circuits (VQCs) and a succinct translation between linguistic structures and quantum circuits. We add two procedures to QNLP that connects its circuits to quantum resources. These are (i) the degree of entanglement of the semantic tensors, and (ii) the degree of contextuality of the quantum system. We experiment with a large dataset designed for a downstream language task: pronominal anaphora resolution and show that whenever there is information flow between a pronoun and a noun, the degrees of entanglement and contextuality of the corresponding circuits increase. This suggests that semantically informative linguistic relations, such as anaphors, are associated with measures of correlations in quantum systems. To our knowledge, this is a connection that has not been established before. The current paper relies on QNLP, which needs syntax to build VQCs. A desirable result would be finding a direct connection between semantic relations, beyond what is syntactically implied.

Download activity - last month

Export as JSONCSVXML

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item