A PennyLane-Centric Dataset to Enhance LLM-based Quantum Code Generation using RAG
Abdul Basit 1 , Nouhaila Innan 1,2 , Muhammad Haider Asif 1,2 , Minghao Shao 1 , Muhammad Kashif 1,2 , Alberto Marchisio 1,2 , Muhammad Shafique 1,2 · 2025
Read A PennyLane-Centric Dataset to Enhance LLM-based Quantum Code Generation using RAG by Abdul Basit 1 , Nouhaila Innan 1,2 , Muhammad Haider Asif 1,2 , Minghao Shao 1 , Muhammad Kashif 1,2 , Alberto Marchisio 1,2 , Muhammad Shafique 1,2 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-02497.txt.
Description
Chat models write fair code in common tongues and stumble in quantum ones, where good samples are scarce. Abdul Basit and six co-authors stock the shelf in 2025 for one quantum tongue, PennyLane. Their set holds near thirty-three hundred code samples with tales of what each does, gleaned from books, docs, and open stores. A build frame kept the gleaning, tagging, and shaping even.
Fed to models through a fetch-first pipe, the set lifted pass rates hard. One small model leapt from under one in ten to over four in ten on quantum tasks. A big model gained less but shed false steps. The team post the set and the frame, and frame both as the fix for niche tongues beyond the quantum one.
Catalog record
| CREATOR | Abdul Basit 1 , Nouhaila Innan 1,2 , Muhammad Haider Asif 1,2 , Minghao Shao 1 , Muhammad Kashif 1,2 , Alberto Marchisio 1,2 , Muhammad Shafique 1,2 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 36 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-02497