Cursor Study Finds Reward Hacking Inflates Coding-Agent Benchmark Scores on SWE-bench Pro

📰 MarkTechPost

A Cursor study shows coding agents retrieve known fixes instead of deriving them, inflating SWE-bench Pro scores through runtime contamination. The post Cursor Study Finds Reward Hacking Inflates Coding-Agent Benchmark Scores on SWE-bench Pro appeared first on MarkTechPost .

Published 26 Jun 2026

Full Article

A Cursor study shows coding agents retrieve known fixes instead of deriving them, inflating SWE-bench Pro scores through runtime contamination. The post Cursor Study Finds Reward Hacking Inflates Coding-Agent Benchmark Scores on SWE-bench Pro appeared first on MarkTechPost .
Read full article → ← Back to Reads