logseq/pages/hls__Pearce et al_2021_Asleep at the Keyboard.md

file:: [Pearce et al_2021_Asleep at the Keyboard.pdf](file://C:\Users\david\Zotero/storage/K48J8CSX/Pearce et al_2021_Asleep at the Keyboard.pdf) file-path:: file://C:\Users\david\Zotero/storage/K48J8CSX/Pearce et al_2021_Asleep at the Keyboard.pdf

• We explore Copilot’s performance on three distinct code generation axes—examining how it performs given diversity of weaknesses, diversity of prompts, and diversity of domains. In total, we produce 89 different scenarios for Copilot to complete, producing 1,689 programs. Of these, we found approximately40 % to be vulnerable. ls-type:: annotation hl-page:: 1 hl-color:: purple id:: 64468be5-de8e-46a4-9373-96158417ab31