# Auditing AI models for Verified Deployment under Semantic Specifications

Source: [Auditing AI models for Verified Deployment under Semantic Specifications](<https://www.sota2.com/research/paper/68fa6b2e1e209131cea2ca24>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T23:42:15.448Z

Authors: Homanga Bharadhwaj, De-An Huang, Chaowei Xiao, Anima Anandkumar, Animesh Garg

Published: 2021-09-25T22:53:24.000Z

## Abstract

Auditing trained deep learning (DL) models prior to deployment is vital for preventing unintended consequences. One of the biggest challenges in auditing is the lack of human-interpretable specifications for the DL models that are directly useful to the auditor. We address this challenge through a sequence of semantically-aligned unit tests, where each unit test verifies whether a predefined specification (e.g., accuracy over 95%) is satisfied with respect to controlled and semantically aligned variations in the input space (e.g., in face recognition, the angle relative to the camera). We enable such unit tests through variations in a semantically-interpretable latent space of a generative model. Further, we conduct certified training for the DL model through a shared latent space representation with the generative model. With evaluations on four different datasets, covering images of chest X-rays, human faces, ImageNet classes, and towers, we show how AuditAI allows us to obtain controlled variations for certified training. Thus, our framework, AuditAI, bridges the gap between semantically-aligned formal verification and scalability. A blog post accompanying the paper is at this link https://developer.nvidia.com/blog/nvidia-research-auditing-ai-models-for-verified-deployment-under-semantic-specifications

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2109.12456>)

[PDF](<https://arxiv.org/pdf/2109.12456.pdf>)
