safety security
The Riskiest Answer Was Not Always the Best One to Review
At a 20 percent review budget, ranking by repair value cut residual wrong-answer exposure from 0.881 to 0.716 in a 720-item stress test.
Summary
At a 20 percent review budget, ranking by repair value cut residual wrong-answer exposure from 0.881 to 0.716 in a 720-item stress test.
The proposed review queue combines estimated wrongness with intervention affordance, impact and cost. On TAT-QA and SciFact items, it barely changed wrong-answer exposure before repair but substantially reduced exposure after deterministic benchmark-supported fixes.
Why it matters
At a 20 percent review budget, ranking by repair value cut residual wrong-answer exposure from 0.881 to 0.716 in a 720-item stress test.
Limits and context
No additional limitation was separately recorded.
Key claims
At a 20 percent review budget, ranking by repair value cut residual wrong-answer exposure from 0.881 to 0.716 in a 720-item stress test.
Evidence: source-2026-09-09-018
Sources
- arXiv preprint 2609.07095arXiv · primary research
Corrections
No corrections have been recorded for this story.