Overview

In the provided evidence, regression tests are mentioned as a downstream use for verification stimuli. A paper on simulation-based processor verification reports that its experiments were able to isolate "a small set of stimuli with high coverage" and that this set "can be used for running regression tests." [C1]

Context in processor verification

The cited work describes a processor-verification workflow in which pseudorandom generators produce stimuli, the stimuli are applied to processor inputs, and functional coverage is monitored to assess verification completeness. [C2]

The same work proposes dynamically altering pseudorandom-generator constraints using a recurrent neural network that receives coverage feedback from simulation of the design under verification. [C3]

Within that workflow, regression tests are not described as the primary generation mechanism; rather, they are presented as a reuse target for a compact, high-coverage stimulus set discovered during experiments. [C1]

Evidence-limited notes

The available evidence does not define the full mechanics of a regression test, such as scheduling, pass/fail criteria, or maintenance policy. It supports only the narrower claim that high-coverage stimuli identified during processor-verification experiments can be used to run regression tests. [C1]