Development
Epoch finds 15% on SWE-bench Verified from prompt restructuring alone
Claim
Epoch AI found a 15% boost on SWE-bench Verified purely from restructuring the prompt.
Reported metric: +15%
Reporting date: (not necessarily the event date)
Sources
Attribution and reporting linked by this record; links do not imply independent verification.
- Welcome to December 28, 2025 (theinnermostloop.substack.com)
Cite or correct this record
Suggest a correction — opens a GitHub draft for you to review and submit.