Is there any plan to use the corrected version of SWE-bench Verified: https://arxiv.org/abs/2506.09289 or https://arxiv.org/abs/2503.15223 ? I'm looking at the SWE-Bench Verified github and don't think these modifications have been incorporated, but feel free to correct me if I'm wrong. The first paper makes it seem significant since the leaderboard changes drastically

Not at the moment, the unfixed version is the standard right now. If anything we might switch to swe-bench pro