ガイド連動クイズ · 中 レベル

SWE-bench and Coding Agent Benchmarks Quiz

Tests how SWE-bench tasks are built, FAIL_TO_PASS and PASS_TO_PASS scoring, the Lite and Verified subsets, and why scaffolds change results.

関連するガイドパスSwe Bench And Coding Agent Benchmarks
質問 1 の 8

What does a SWE-bench task ask the system to do?