TAPAS: Datasets for Learning the Learning with Errors Problem

Paper 2025/1968

TAPAS: Datasets for Learning the Learning with Errors Problem

Eshika Saxena

, FAIR at Meta

Alberto Alfarano, FAIR at Meta

François Charton

Emily Wenger

, Duke University

Kristin Lauter

, FAIR at Meta

Abstract

AI-powered attacks on Learning with Errors (LWE), an important hard math problem in post-quantum cryptography, rival or outperform "classical" attacks on LWE under certain parameter settings. Despite the promise of this approach, a dearth of accessible data limits AI practitioners' ability to study and improve these attacks. Creating LWE data for AI model training is time- and compute-intensive and requires significant domain expertise. To fill this gap and accelerate AI research on LWE attacks, we propose the TAPAS datasets, a Toolkit for Analysis of Post-quantum cryptography using AI Systems. These datasets cover several LWE settings and can be used off-the-shelf by AI practitioners to prototype new approaches to cracking LWE. This work documents TAPAS dataset creation, establishes attack performance baselines, and lays out directions for future work.

Metadata

Available format(s): PDF
Category: Attacks and cryptanalysis
Publication info: Published elsewhere. NeurIPS Datasets and Benchmarks 2025
Keywords: Learning with Errors Cryptanalysis Machine Learning
Contact author(s): eshika @ meta com
emily wenger @ duke edu
klauter @ meta com
History: 2025-12-19: revised; 2025-10-20: received; See all versions
Short URL: https://ia.cr/2025/1968
License: CC BY

BibTeX

@misc{cryptoeprint:2025/1968,
      author = {Eshika Saxena and Alberto Alfarano and François Charton and Emily Wenger and Kristin Lauter},
      title = {{TAPAS}: Datasets for Learning the Learning with Errors Problem},
      howpublished = {Cryptology {ePrint} Archive, Paper 2025/1968},
      year = {2025},
      url = {https://eprint.iacr.org/2025/1968}
}