ResearchBenchmarksAutomated agent rollback on GSM8K Left Dense variant (test)Automated agent rollback on GSM8K Left Dense variant (test)Follow77Success RateLF72.8473.927576.08May 7, 2026Evaluation resultsMethodMethodLinksSuccess RateCoverageCostLFScoring function=fine-...Scoring function=fine-tuned LLM, Target coverage=80%2026.0577810.9Top-1Scoring function=fine-...Scoring function=fine-tuned LLM2026.0576920.83VCPScoring function=fine-...Scoring function=fine-tuned LLM, Target coverage=80%2026.0573850.79