ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2506.09455
69
0

Abstraction-Based Proof Production in Formal Verification of Neural Networks

11 June 2025
Y. Elboher
Omri Isac
Guy Katz
Tobias Ladner
Haoze Wu
ArXiv (abs)PDFHTML
Main:13 Pages
7 Figures
Bibliography:4 Pages
Abstract

Modern verification tools for deep neural networks (DNNs) increasingly rely on abstraction to scale to realistic architectures. In parallel, proof production is becoming a critical requirement for increasing the reliability of DNN verification results. However, current proofproducing verifiers do not support abstraction-based reasoning, creating a gap between scalability and provable guarantees. We address this gap by introducing a novel framework for proof-producing abstraction-based DNN verification. Our approach modularly separates the verification task into two components: (i) proving the correctness of an abstract network, and (ii) proving the soundness of the abstraction with respect to the original DNN. The former can be handled by existing proof-producing verifiers, whereas we propose the first method for generating formal proofs for the latter. This preliminary work aims to enable scalable and trustworthy verification by supporting common abstraction techniques within a formal proof framework.

View on arXiv
@article{elboher2025_2506.09455,
  title={ Abstraction-Based Proof Production in Formal Verification of Neural Networks },
  author={ Yizhak Yisrael Elboher and Omri Isac and Guy Katz and Tobias Ladner and Haoze Wu },
  journal={arXiv preprint arXiv:2506.09455},
  year={ 2025 }
}
Comments on this paper