AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Break-It-Fix-It: Unsupervised Learning for Program Repair

Michihiro Yasunaga, Percy Liang

ICML 2021 · Label-efficient and robust learning › Recognition and long-tail · approach: Data augmentation and generalization bias · 27 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use BIFI: add critic-verified fixer outputs on real bad code and breaker-generated bad examplesInitial fixer on GitHub-Python · GitHub-Python test setrepair accuracy: 62.0% -> 90.5%Improved
Use BIFI on the DrRepair initializationDrRepair · DeepFix test setrepair accuracy: 66.1% -> 71.7%Improved
Add real bad examples to fixer training without a learned breakerInitial fixer · GitHub-Python test setrepair accuracy: 62.0% -> 88.6%Improved
Add breaker-generated bad examples alongside real bad examplesFixerOnly on GitHub-Python · GitHub-Python test setrepair accuracy: 88.6% -> 90.5%Improved
Add breaker-generated bad examples when real bad data is limitedFixerOnly on GitHub-Python · GitHub-Python, Bad 10%repair accuracy: 78.5% -> 86.7%Improved
Train on a small amount of real bad examples instead of relying on synthetic bad examples onlyFixerOnly · GitHub-Python, Bad 10%repair accuracy: 62.7% -> 78.5%Improved
Remove real bad examples from fixer training (ablation)BIFI · GitHub-Python test setrepair accuracy: 90.5% -> 84.6%Hurt
Remove critic verification of fix and break attempts (ablation)BIFI · GitHub-Python test setrepair accuracy: 90.5% -> 84.0%Hurt
Remove both critic verification and real bad examples, as in backtranslation (ablation)BIFI · GitHub-Python test setrepair accuracy: 90.5% -> 80.1%Hurt
Continue training the fixer using synthetic perturbations only (ablation)Initial fixer · GitHub-Python test setrepair accuracyNo clear effect
Reduce available real bad examples from Bad 100% to Bad 50% (ablation)BIFI · GitHub-Python test setrepair accuracy: 90.5% -> 89.0%Hurt
Reduce available real bad examples from Bad 100% to Bad 10% (ablation)BIFI · GitHub-Python test setrepair accuracy: 90.5% -> 86.7%Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy