Quantum Compositional NLP for Arabic: Grammar, Morphology, and Word Sense in Circuit Topology

Short summary

This paper presents the first application of pregroup grammar-based quantum compositional NLP (QNLP) to Arabic, a morphologically rich free-word-order language. Arabic sentences are converted into quantum circuits where grammatical structure determines gate topology. Three experiments cover word order, morphological tense, and verb sense disambiguation, comparing quantum circuit methods against classical baselines including AraVec embeddings and AraBERT. The work tests whether quantum circuit topology can effectively model Arabic's complex compositional semantics.

• •First QNLP application to Arabic using pregroup grammar-based quantum circuits
• •Three experiments: word order, tense, and verb sense disambiguation
• •Compared against AraVec and AraBERT classical baselines

Generated with AI, which can make mistakes.