@ -10,9 +10,9 @@ The key findings of their prompt engineering approach are:
- The impact of the prompt on eliciting the correct reasoning is massive. Simply asking the model to classify a given job results in an F1 score of 65.6, whereas the post-prompt engineering model achieves an F1 score of 91.7.
- Attempting to force the model to stick to a template lowers performance in all cases (this behaviour disappears in early testing with GPT-4, which are posterior to the paper).
- Many small modifications have an outsized impact on performance.
- The tables below show the full modifications tested.
- Properly giving instructions and repeating the key points appears to be the biggest performance driver.
- Something as simple as giving the model a (human) name and referring to it as such increased F1 score by 0.6pts.
- The tables below show the full modifications tested.
- Properly giving instructions and repeating the key points appears to be the biggest performance driver.
- Something as simple as giving the model a (human) name and referring to it as such increased F1 score by 0.6pts.
### Prompt Modifications Tested
@ -53,4 +53,4 @@ The key findings of their prompt engineering approach are: