Research

Paper

TESTING March 10, 2026

Avoiding Big Integers: Parallel Multimodular Algebraic Verification of Arithmetic Circuits

Authors

Clemens Hofstadler, Daniela Kaufmann, Chen Chen

Abstract

Word-level verification of arithmetic circuits with large operands typically relies on arbitrary-precision arithmetic, which can lead to significant computational overhead as word sizes grow. In this paper, we present a hybrid algebraic verification technique based on polynomial reasoning that combines linear and nonlinear rewriting. Our approach relies on multimodular reasoning using homomorphic images, where computations are performed in parallel modulo different primes, thereby avoiding any large-integer arithmetic. We implement the proposed method in the verification tool TalisMan2.0 and evaluate it on a suite of multiplier benchmarks. Our results show that hybrid multimodular reasoning significantly improves upon existing approaches.

Metadata

arXiv ID: 2603.09501
Provider: ARXIV
Primary Category: cs.SC
Published: 2026-03-10
Fetched: 2026-03-11 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.09501v1</id>\n    <title>Avoiding Big Integers: Parallel Multimodular Algebraic Verification of Arithmetic Circuits</title>\n    <updated>2026-03-10T11:05:47Z</updated>\n    <link href='https://arxiv.org/abs/2603.09501v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.09501v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Word-level verification of arithmetic circuits with large operands typically relies on arbitrary-precision arithmetic, which can lead to significant computational overhead as word sizes grow. In this paper, we present a hybrid algebraic verification technique based on polynomial reasoning that combines linear and nonlinear rewriting. Our approach relies on multimodular reasoning using homomorphic images, where computations are performed in parallel modulo different primes, thereby avoiding any large-integer arithmetic. We implement the proposed method in the verification tool TalisMan2.0 and evaluate it on a suite of multiplier benchmarks. Our results show that hybrid multimodular reasoning significantly improves upon existing approaches.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.SC'/>\n    <published>2026-03-10T11:05:47Z</published>\n    <arxiv:comment>Submitted to IJCAR 2026</arxiv:comment>\n    <arxiv:primary_category term='cs.SC'/>\n    <author>\n      <name>Clemens Hofstadler</name>\n    </author>\n    <author>\n      <name>Daniela Kaufmann</name>\n    </author>\n    <author>\n      <name>Chen Chen</name>\n    </author>\n  </entry>"
}