The goal of this module, related to the specific objectives OB2 and OB3, is twofold. On the one hand, we aim to determine how quantum theory could be integrated into classical algorithms, for instance, for optimisation purposes, resulting in hybrid classical-quantum algorithms. On the other hand, we would like to explore novel quantum or quantum-inspired algorithms or improve existing ones. In both cases, the developed algorithms would be later applied to solve NLP tasks and applications and the used case defined in modules 3 and 4, respectively (M3 and M4).
Algorithms at the intersection of NLP and quantum physics can be implemented either on quantum computers or classical computers. The former ones are usually called quantum algorithms, and the latter ones are usually named quantum-inspired models (Montanaro, 2016). Both types of algorithms can model different features in the language. In this context, we find several lines of research related to Quantum Machine Learning (Biamonte et al., 2017; Peral-García, Cruz-Benito, and García-Peñalvo, 2024b), Quantum Deep Learning (Alchieri et al., 2021) or Quantum-Inspired Machine Learning (Huynh et al., 2023), this latter focusing on developing classical machine learning algorithms inspired by principles of quantum mechanics within a classical computational framework. These include algorithms, such as Quantum Support Vector Machine (Simões et al., 2023), Quantum Convolutional Neural Network (Hur, Kim and Park, 2022), Quantum Recurrent Neural Network (Yu et al., 2024), Quantum Generative Adversarial Networks (Tian et al., 2023) or Quantum-Inspired language models (Fan et al., 2024), among others, that have been shown superior accuracy over their classical counterparts.
Moreover, applying quantum mechanics concepts to describe features enhances interpretability due to their transparent physical explanations (Wu et al., 2021). It is also more beneficial to the subsequent neural network to extract useful information, as it has been proposed through the QIXAI framework, a novel approach for enhancing neural network interpretability through quantum-inspired mathematical methods (Willis, 2024).
Another interesting application of quantum algorithms is directly related to optimisation to achieve faster and more effective results (Blekos et al., 2024). This could include novel approaches in which quantum-based solvers may be used in tasks related to feature selection (Mandal and Chakraborty, 2024), relevant for NLP applications, such as information retrieval, text classification, summarisation, or simplification. The QuantumCLEF shared tasks (Pasin et al., 2024), proposed for the CLEF 2024 and CLEF 2025 conferences, provide a promising starting point to analyse existing approaches in this area and address the issue.