swe-infinite: score unsubmitted miniswe runs as zero#73
Open
catoneone wants to merge 1 commit into
Open
Conversation
Take the patch only from a Submitted run; LimitsExceeded and other terminating statuses now yield an empty patch with no container git-diff fallback, matching upstream mini-swe-agent + SWE-bench semantics. Step-limit runs are recorded as failure_reason limit_reached_no_submission.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Submittedrun inMiniSWEAgent.solve. Any other terminating status (e.g.LimitsExceededat the step limit) yields an empty patch — deliberately no container git-diff fallback — matching upstream mini-swe-agent + SWE-bench semantics: no submission = empty patch = score 0.failure_reason: limit_reached_no_submissioninenv.pyfor observability (distinct fromno_patch_generated).exit_statustoMiniSWEResult.Production data (2026-06-30 ~ 07-14, 227 miniswe rollouts): 22.5% hit the step limit without submitting; the diff-harvest fallback in the deployed image let 13 unsubmitted runs score 1.0.
Codex/affent are unchanged: they have no submission protocol, so post-run diff extraction remains their normal patch path.
Test
test_submission_gating.py: 3/3 pass (submitted patch accepted; step-limit and non-submit terminating messages yield empty patch).test_structured_outcomes.py: 21 passed.