Skip to content

bench(reviewer-eval): waive hunk drops for rebutted candidates — decoys don't need the hunk #306

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
norvalbv merged 1 commit into from
Aug 2, 2026
+4 −0
Merged

bench(reviewer-eval): waive hunk drops for rebutted candidates — decoys don't need the hunk #306

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions gate-engine/review/eval/reviewers/propose.mts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -74,6 +74,10 @@ function hardDropReason(c) {
if (!c.originalCommitId) return 'missing-original-commit';
if (c.line === null && c.originalLine === null) return 'no-line';
if (c.outcomeEvidence === 'outdated-only') return 'outdated-only';
// Rebutted candidates mint DECOYS — the scarce corpus resource. Their fixture is built from
// baseFileContent + the bot's (rebutted) claim, not from the diff hunk, so hunk absence/size
// must not drop them. Golds (fixed) still need the hunk to locate the defect being fixed.
if (c.outcome === 'rebutted') return null;
if (!c.diffHunk || c.diffHunk.length === 0) return 'empty-hunk';
if ((c.hunkLen ?? c.diffHunk.length) > MAX_HUNK_LEN) return 'hunk-too-long';
return null;
Expand Down
Loading