CogBattery: Human-Referenced Evaluation of Cognitive Task Behavior and Structure in Language Models

Research Hugging Face Papers Source language: Chinese

论文摘要(作者报告,未经本站复测):Human cognitive tasks let us compare language models with humans under controlled conditions. Yet a model can reproduce the direction of a human effect while making different errors.

Excerpt

论文摘要(作者报告,未经本站复测):Human cognitive tasks let us compare language models with humans under controlled conditions. Yet a model can reproduce the direction of a human effect while making different errors.

Short excerpt from the source — open the original for the full text.

Sources

← Back to the daily brief