LLM Evaluation
The methodology and engineering practice for assessing the output quality of large language models. Common approaches include model-based mutual evaluation and human-annotated datasets to measure accuracy, safety, and consistency. It is a core component of quality assurance in AI systems.
Timeline (last 90 days)
LLM Eval typically relies on a model-evaluates-model approach, or uses human-annotated golden datasets to measure output accuracy, safety, and stylistic consistency.
LLM evaluation needs to simultaneously consider multiple dimensions such as factual accuracy, instruction-following, harmlessness, helpfulness, and expression quality, and tensions among these dimensions are common.
All Facts (2)
LLM evaluation needs to simultaneously consider multiple dimensions such as factual accuracy, instruction-following, harmlessness, helpfulness, and expression quality, and tensions among these dimensions are common.
60%UnverifiedLLM Eval typically relies on a model-evaluates-model approach, or uses human-annotated golden datasets to measure output accuracy, safety, and stylistic consistency.
50%