STORY · FORSKNING_

Language models reward domain expertise

An analysis of how domain knowledge affects the ability to extract value from large language models. The author uses Terence Tao's mathematical conversations with ChatGPT as an example of how professional expertise provides greater control over the model's output than pure prompting techniques.

WHY IT MATTERS

The piece argues that human expertise will remain valuable as models become more powerful, because the difficult part is communicating to the model what people want – not the model's capabilities.

SOURCES

MACHINE-GENERATED SUMMARY This summary is written by machine from the sources below. We sort and explain — but we are a way into the field, not the final word. Check the source when something matters to you.