When Models Edit Too Much: On the Fidelity of Minimal Code Edits
A 400-task BigCodeBench evaluation shows frontier LLMs widely over-edit code; a preservation instruction cuts excess edits and raises Pass@1 by 2.3 points.
1 stories, newest first · one card per story
A 400-task BigCodeBench evaluation shows frontier LLMs widely over-edit code; a preservation instruction cuts excess edits and raises Pass@1 by 2.3 points.
Today 1.6K readers · 2.5K views · 779 reads · all time 189K views