/TECHLessWrong Analysis Flags Flaws in OpenAI's New SWE-Bench ProTRNT#1704Combined views3313 posts, first seen 3h ago3 likes1 commentsLessWrong Analysis Flags Flaws in OpenAI's New SWE-Bench ProTRNT#1704Source Posts on XCluster sourcesTalia Ringer 🕊🪬@TaliaRingerTECHTerrible https://x.com/TaliaRinger/status/2075233130642919478?s=20Combined views3313 posts, first seen 3h ago3 likes1 comments