Comparative Study of the Ansätze in Quantum Language Models

Abstract Quantum language models are the alternative to classical language models, which borrow concepts and methods from quantum machine learning and computational linguistics. While several quantum natural language processing (QNLP) methods and frameworks exist for text classification and generation, there is a lack of systematic study to compare the performance across various ansätze, in terms of their hyperparameters and classical and quantum methods to implement them. Here, the performance of quantum natural language processing models based on these ansätze is evaluated at different levels in text classification tasks. A comparative study is performed and the QNLP models are optimized by fine‐tuning several critical hyperparameters. These results demonstrate how the balance between simplification and expressivity affects model performance. This study provides extensive data to improve the understanding of QNLP models and opens the possibility of developing better QNLP algorithms.