Understanding and Meeting Practitioner Needs When Measuring Representational Harms Caused by LLM-Based Systems

Open in new window