Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

International Conference on Multidisciplinary Research and Innovations in Engineering (MRIE)

A Hybrid Quantum-Classical Language Model for Text Generation: A Case Study on The Gift of the Magi

Aadrian Routh, Juhi M Singh

Published Jul 30, 2025• 1 citation
Publisher

Abstract

This paper presents a hybrid quantum-classical language model for text generation, evaluated through a detailed case study of O. Henry's classic short story, The Gift of the Magi. The model integrates a classical transformer with a quantum generator to enhance the extraction of abstract narrative features. Our system architecture includes stages for input tokenization, contextual embedding through multi-head attention, quantumenhanced feature extraction using a parameterized quantum circuit, and output refinement through a classical discriminator. Quantitative evaluations using the BLEU score, besides qualitative analysis, reveal that while the hybrid approach generates syntactically coherent text, it currently falls short in capturing the nuanced symbolic and emotional depth of the original narrative. The findings show the present limitations of quantum-enhanced language models and point towards future research directions in refining quantum circuit designs, reducing noise, and improving integration strategies to fully exploit the potential of quantum computing in natural language processing.

DOI
10.1109/mrie66930.2025.11156371

Related papers

Quantum-Enhanced Natural Language Generation: A Multi-Model Framework with Hybrid Quantum-Classical Architectures

Chi-Sheng Chen, En-Jui Kuo

This paper presents a comprehensive evaluation of quantum text generation models against traditional Transformer/MLP architectures, addressing the growing interest in quantum computing applications for natural language processing. We conduct systematic experiments comparing five distinct models: Transformer (baseline), Quantum Kernel Self-Attention Network (QKSAN), Quantum RWKV (QRWKV), and Quantum Attention Sequence Architecture (QASA) across five diverse datasets including simple sentences, short stories, quantum phrases, haiku poetry, and proverbs. Our evaluation employs multiple metrics including perplexity, BLEU scores, vocabulary diversity, repetition rates, and fluency measures to assess different aspects of text generation quality. The experimental results reveal that while traditional Transformer models maintain overall superiority with the lowest average perplexity (1.21) and highest BLEU-1 score (0.2895), quantum-inspired models demonstrate competitive performance in specific scenarios. Notably, QKSAN achieves a competitive BLEU-1 score of 0.2800 while maintaining zero repetition rates, and QRWKV demonstrates perfect vocabulary diversity (Distinct-1 = 1.000) in certain tasks.

arXiv:2508.21332•Aug 29, 2025•0 citations

Quantum-Enhanced Attention Mechanism in NLP: A Hybrid Classical-Quantum Approach

S. M. Yousuf Iqbal Tomal, Abdullah Al Shafin, Bhattacharjee, Debojit +2 more

Recent advances in quantum computing have opened new pathways for enhancing deep learning architectures, particularly in domains characterized by high-dimensional and context-rich data such as natural language processing (NLP). In this work, we present a hybrid classical-quantum Transformer model that integrates a quantum-enhanced attention mechanism into the standard classical architecture. By embedding token representations into a quantum Hilbert space via parameterized variational circuits and exploiting entanglement-aware kernel similarities, the model captures complex semantic relationships beyond the reach of conventional dot-product attention. We demonstrate the effectiveness of this approach across diverse NLP benchmarks, showing improvements in both efficiency and representational capacity. The results section reveal that the quantum attention layer yields globally coherent attention maps and more separable latent features, while requiring comparatively fewer parameters than classical counterparts. These findings highlight the potential of quantum-classical hybrid models to serve as a powerful and resource-efficient alternative to existing attention mechanisms in NLP.

arXiv:2501.15630•Jan 26, 2025•3 citations

QuantumGPTMini: A Hybrid Quantum-Classical Transformer for Enhanced NLP

Ramazan Amire, Amire Ramazan

Transformer architectures have driven advances in natural language processing, powering systems from question answering to creative text generation. Yet as model sizes have grown from hundreds of millions to hundreds of billions of parameters, the computational cost of these systems has become prohibitive, placing advanced language technology out of reach for many researchers. In this work, we introduce QuantumGPTMini, a hybrid quantum-classical transformer with just over one million classical parameters that leverages lightweight quantum circuits to perform its core attention and feed-forward computations. By encoding token embeddings both in conventional vectors and in the amplitudes of small variational quantum circuits, our model replaces the conventional softmax attention mechanism with an entanglement-driven alternative. When trained on the Wikipedia-2 corpus using PennyLane's lightning.qubit simulator on NVIDIA A100 GPUs, QuantumGPTMini matched the perplexity of a GPT-2 Small baseline while demonstrating superior pattern recognition and symbolic reasoning behaviors. We have made this system publicly accessible through a FastAPI endpoint and a Next.js web interface. While our initial results underscore the potential of quantum-augmented language models, they also highlight challenges in moving beyond simulation, including circuit depth limitations, decoherence, and the high overhead of parameter-shift gradient estimation.

•Apr 28, 2025•0 citations

Quantum Natural Language Processing Using Hybrid AI Algorithms: A Variational Quantum–Classical Framework for Efficient Text Classification

P. Anitha

Quantum Natural Language Processing (QNLP) represents a novel interdisciplinaryparadigm that integrates Quantum Computing with Natural Language Processing toenhance semantic modeling and computational efficiency. Despite the remarkable successof transformer-based architectures such as BERT and GPT, classical NLP systems requirelarge-scale computational resources and extensive parameter tuning. This paper proposesa hybrid quantum–classical architecture employing parameterized quantum circuits (PQCs)integrated with classical embedding layers for text classification tasks. The frameworkutilizes classical preprocessing, quantum state encoding, variational optimization, andclassical post-measurement classification. Experimental results on benchmark sentimentanalysis datasets demonstrate competitive accuracy with significantly reduced parametercomplexity. The findings indicate that hybrid QNLP models provide enhanced semanticcompositionality and efficient representation in Hilbert space while remaining feasiblewithin Noisy Intermediate-Scale Quantum (NISQ) constraints

International Journal For Multidisciplinary Research•Mar 4, 2026•0 citations

Hybrid Quantum-Classical NLP Framework for Emotion Recognition Tasks

Dharaniraj VM, Jayasuriya R, Sivasankaramoorthy R +2 more

This project aims at developing a comprehensive hybrid quantum-classical NLP pipeline to solve the task of emotion recognition. For emotion recognition, this work makes use of a hybrid Quantum Recurrent Neural Network (QRNN) by combining classical embedding layers with parameterized quantum circuits based on well defined ansatzes. The primary goal of the project is to train VQCs for capturing the emotion associated with sentences. This work aims at combining the strong embedding generation ability of classical architectures with the expressive power of Quantum circuits that can capture complex dependencies across sequences. The designed Hybrid QRNN model integrates two important aspects namely the strength of RNNs for sequential modelling and the representational power of quantum circuits resulting in competitive performance for emotion recognition task while also utilising the quantum benefits. The results reveal the potential of QNLP for advancing the intersection of Quantum Computing and Natural Language Processing for quantum-accelerated downstream NLP applications.

EPJ Web of Conferences•Jan 1, 2026•0 citations

Hybrid Quantum-Classical NLP Framework for Emotion Recognition Tasks

Dharaniraj VM, Jayasuriya R, Sivasankaramoorthy R +2 more

This project aims at developing a comprehensive hybrid quantum-classical NLP pipeline to solve the task of emotion recognition. For emotion recognition, this work makes use of a hybrid Quantum Recurrent Neural Network (QRNN) by combining classical embedding layers with parameterized quantum circuits based on well defined ansatzes. The primary goal of the project is to train VQCs for capturing the emotion associated with sentences. This work aims at combining the strong embedding generation ability of classical architectures with the expressive power of Quantum circuits that can capture complex dependencies across sequences. The designed Hybrid QRNN model integrates two important aspects namely the strength of RNNs for sequential modelling and the representational power of quantum circuits resulting in competitive performance for emotion recognition task while also utilising the quantum benefits. The results reveal the potential of QNLP for advancing the intersection of Quantum Computing and Natural Language Processing for quantum-accelerated downstream NLP applications.

•Mar 23, 2026•0 citations
This paper presents a hybrid quantum-classical language model for text generation, evaluated through a detailed case study of O. Henry's classic short story, The Gift of the Magi. The model integrates a classical transformer with a quantum generator to enhance the extraction of abstract narrative features. Our system architecture includes stages for input tokenization, contextual embedding through multi-head attention, quantumenhanced feature extraction using a parameterized quantum circuit, and output refinement through a classical discriminator. Quantitative evaluations using the BLEU score, besides qualitative analysis, reveal that while the hybrid approach generates syntactically coherent text, it currently falls short in capturing the nuanced symbolic and emotional depth of the original narrative. The findings show the present limitations of quantum-enhanced language models and point towards future research directions in refining quantum circuit designs, reducing noise, and improving integration strategies to fully exploit the potential of quantum computing in natural language processing.
This paper presents a comprehensive evaluation of quantum text generation models against traditional Transformer/MLP architectures, addressing the growing interest in quantum computing applications for natural language processing. We conduct systematic experiments comparing five distinct models: Transformer (baseline), Quantum Kernel Self-Attention Network (QKSAN), Quantum RWKV (QRWKV), and Quantum Attention Sequence Architecture (QASA) across five diverse datasets including simple sentences, short stories, quantum phrases, haiku poetry, and proverbs. Our evaluation employs multiple metrics including perplexity, BLEU scores, vocabulary diversity, repetition rates, and fluency measures to assess different aspects of text generation quality. The experimental results reveal that while traditional Transformer models maintain overall superiority with the lowest average perplexity (1.21) and highest BLEU-1 score (0.2895), quantum-inspired models demonstrate competitive performance in specific scenarios. Notably, QKSAN achieves a competitive BLEU-1 score of 0.2800 while maintaining zero repetition rates, and QRWKV demonstrates perfect vocabulary diversity (Distinct-1 = 1.000) in certain tasks.
Recent advances in quantum computing have opened new pathways for enhancing deep learning architectures, particularly in domains characterized by high-dimensional and context-rich data such as natural language processing (NLP). In this work, we present a hybrid classical-quantum Transformer model that integrates a quantum-enhanced attention mechanism into the standard classical architecture. By embedding token representations into a quantum Hilbert space via parameterized variational circuits and exploiting entanglement-aware kernel similarities, the model captures complex semantic relationships beyond the reach of conventional dot-product attention. We demonstrate the effectiveness of this approach across diverse NLP benchmarks, showing improvements in both efficiency and representational capacity. The results section reveal that the quantum attention layer yields globally coherent attention maps and more separable latent features, while requiring comparatively fewer parameters than classical counterparts. These findings highlight the potential of quantum-classical hybrid models to serve as a powerful and resource-efficient alternative to existing attention mechanisms in NLP.
Transformer architectures have driven advances in natural language processing, powering systems from question answering to creative text generation. Yet as model sizes have grown from hundreds of millions to hundreds of billions of parameters, the computational cost of these systems has become prohibitive, placing advanced language technology out of reach for many researchers. In this work, we introduce QuantumGPTMini, a hybrid quantum-classical transformer with just over one million classical parameters that leverages lightweight quantum circuits to perform its core attention and feed-forward computations. By encoding token embeddings both in conventional vectors and in the amplitudes of small variational quantum circuits, our model replaces the conventional softmax attention mechanism with an entanglement-driven alternative. When trained on the Wikipedia-2 corpus using PennyLane's lightning.qubit simulator on NVIDIA A100 GPUs, QuantumGPTMini matched the perplexity of a GPT-2 Small baseline while demonstrating superior pattern recognition and symbolic reasoning behaviors. We have made this system publicly accessible through a FastAPI endpoint and a Next.js web interface. While our initial results underscore the potential of quantum-augmented language models, they also highlight challenges in moving beyond simulation, including circuit depth limitations, decoherence, and the high overhead of parameter-shift gradient estimation.
Quantum Natural Language Processing (QNLP) represents a novel interdisciplinaryparadigm that integrates Quantum Computing with Natural Language Processing toenhance semantic modeling and computational efficiency. Despite the remarkable successof transformer-based architectures such as BERT and GPT, classical NLP systems requirelarge-scale computational resources and extensive parameter tuning. This paper proposesa hybrid quantum–classical architecture employing parameterized quantum circuits (PQCs)integrated with classical embedding layers for text classification tasks. The frameworkutilizes classical preprocessing, quantum state encoding, variational optimization, andclassical post-measurement classification. Experimental results on benchmark sentimentanalysis datasets demonstrate competitive accuracy with significantly reduced parametercomplexity. The findings indicate that hybrid QNLP models provide enhanced semanticcompositionality and efficient representation in Hilbert space while remaining feasiblewithin Noisy Intermediate-Scale Quantum (NISQ) constraints
This project aims at developing a comprehensive hybrid quantum-classical NLP pipeline to solve the task of emotion recognition. For emotion recognition, this work makes use of a hybrid Quantum Recurrent Neural Network (QRNN) by combining classical embedding layers with parameterized quantum circuits based on well defined ansatzes. The primary goal of the project is to train VQCs for capturing the emotion associated with sentences. This work aims at combining the strong embedding generation ability of classical architectures with the expressive power of Quantum circuits that can capture complex dependencies across sequences. The designed Hybrid QRNN model integrates two important aspects namely the strength of RNNs for sequential modelling and the representational power of quantum circuits resulting in competitive performance for emotion recognition task while also utilising the quantum benefits. The results reveal the potential of QNLP for advancing the intersection of Quantum Computing and Natural Language Processing for quantum-accelerated downstream NLP applications.
This project aims at developing a comprehensive hybrid quantum-classical NLP pipeline to solve the task of emotion recognition. For emotion recognition, this work makes use of a hybrid Quantum Recurrent Neural Network (QRNN) by combining classical embedding layers with parameterized quantum circuits based on well defined ansatzes. The primary goal of the project is to train VQCs for capturing the emotion associated with sentences. This work aims at combining the strong embedding generation ability of classical architectures with the expressive power of Quantum circuits that can capture complex dependencies across sequences. The designed Hybrid QRNN model integrates two important aspects namely the strength of RNNs for sequential modelling and the representational power of quantum circuits resulting in competitive performance for emotion recognition task while also utilising the quantum benefits. The results reveal the potential of QNLP for advancing the intersection of Quantum Computing and Natural Language Processing for quantum-accelerated downstream NLP applications.