Scaling Hindi Quantum Natural Language Processing through Automatic Pregroup Supertagging
Abstract
Quantum Natural Language Processing (QNLP) uses pregroup grammars to translate grammatical structure into diagrammatic representations and quantum circuits. Recent Hindi QNLP work has shown that Hindi-specific pregroup grammars can support grammar-sensitive compositional models, but grammatical type assignment is still largely manual, limiting scalability. This paper formulates automatic Hindi pregroup supertagging as a token-level classification task. Using a manually annotated corpus of 380 Hindi sentences, we evaluate lexical, contextual, prompting-based, lexical-repair, and suffix/morphology-aware methods. Results show that simple lexical and contextual models are strong in this low-resource setting: contextual backoff achieves the best completed accuracy of 64.56\%, while raw Qwen2.5 prompting reaches only 11.65\%. Lexical repair raises LLM-assisted prediction to 64.08\%, demonstrating the value of constraining generative outputs with symbolic grammar knowledge. Diagnostic analysis further shows that seen and unambiguous tokens are much easier than unseen tokens, and suffix/morphology features improve karaka-token accuracy but not overall performance. These results show that automatic Hindi pregroup assignment is feasible and can reduce reliance on manual annotation in future multilingual QNLP pipelines.
Related
- Scaling the Wavelength to Optimize Ultrafast Laser Processing of Bandgap Materials
- Overexposure-free automatic exposure algorithm for division-of-aperture polarization camera
- Penalized PET/CT Reconstruction Algorithms With Automatic Realignment for Anatomical Priors
- Derivative Computation in PINNs: Automatic Differentiation, Finite Differences and Beyond
- Automatic control of gold nanoparticle clustering in plasmonic nanopores based on Raman spectroscopic feedback
BibTeX
@misc{naik2026scaling,
title = {{Scaling Hindi Quantum Natural Language Processing through Automatic Pregroup Supertagging}},
author = {Gautami Sanjay Naik and Krishna Bhatia and Mithun Paul Saint-Germain and H Aswath Babu},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.13721},
eprint = {2609.13721},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.13721},
}
RIS
TY - UNPB
TI - Scaling Hindi Quantum Natural Language Processing through Automatic Pregroup Supertagging
AU - Gautami Sanjay Naik
AU - Krishna Bhatia
AU - Mithun Paul Saint-Germain
AU - H Aswath Babu
PY - 2026
DA - 2026/09/12
PB - arXiv
DO - 10.48550/arxiv.2609.13721
UR - https://arxiv.org/abs/2609.13721
AB - Quantum Natural Language Processing (QNLP) uses pregroup grammars to translate grammatical structure into diagrammatic representations and quantum circuits. Recent Hindi QNLP work has shown that Hindi-specific pregroup grammars can support grammar-sensitive compositional models, but grammatical type assignment is still largely manual, limiting scalability. This paper formulates automatic Hindi pregroup supertagging as a token-level classification task. Using a manually annotated corpus of 380 Hindi sentences, we evaluate lexical, contextual, prompting-based, lexical-repair, and suffix/morphology-aware methods. Results show that simple lexical and contextual models are strong in this low-resource setting: contextual backoff achieves the best completed accuracy of 64.56\%, while raw Qwen2.5 prompting reaches only 11.65\%. Lexical repair raises LLM-assisted prediction to 64.08\%, demonstrating the value of constraining generative outputs with symbolic grammar knowledge. Diagnostic analysis further shows that seen and unambiguous tokens are much easier than unseen tokens, and suffix/morphology features improve karaka-token accuracy but not overall performance. These results show that automatic Hindi pregroup assignment is feasible and can reduce reliance on manual annotation in future multilingual QNLP pipelines.
ER -