ProgramBench Vetted: Reverse Engineering from a Runnable Binaryvetto.ai20 points by rigelbm 2 days ago
GIULIAFEDERIGHI 2 days agoA benchmark can be difficult for the wrong reasons and the interesting part here is making sure the score reflects actual reverse-engineering ability not missing information or shortcuts.
A benchmark can be difficult for the wrong reasons and the interesting part here is making sure the score reflects actual reverse-engineering ability not missing information or shortcuts.