Refusal
SDK Usage
Learn how to evaluate this metric programmatically in the Trustwise SDK Documentation.
The Trustwise Refusal metric measures how clearly and firmly a response declines to answer a query. It gives higher scores when the response directly refuses without engaging with content. The metric evaluates both the question and response.
FAQs
Why is Refusal evaluation necessary?
Agents without proper refusal mechanisms may generate harmful or non-compliant outputs, putting users or organizations at risk.
What does a high Refusal score mean?
It indicates that the agent correctly recognized and declined to answer an inappropriate or harmful query.