Do Prompt-Based Models Really Understand the Meaning of thei...

Do Prompt-Based Models Really Understand the Meaning of their Prompts?

RelatedInsightsCollectionsHighlights

Holy shit... Stanford just published research that destroys every prompt engineering guru. Turns out most "advanced techniques" are just survivorship bias and confirmation bias. Here's what the data actually shows:

Alex Hughes x.com

i have noticed that LLMs like claude and gpt-4o work really well with this prompt, it instructs them to 'contemplate' for a bit before giving the final answer. https://t.co/9o5Q9M2Atv

maharshi

x.com

Large Language Models Are Human-Level Prompt Engineers abs: https://t.co/eqNfe2o4JQ project page: https://t.co/tusmJehApu https://t.co/Vw7UU36XJF

x.com