AI Paper Atlas

Research map / Label-efficient and robust learning / OOD and anomaly detection

On Distribution Shift in Learning-based Bug Detectors

Jingxuan He, Luca Beurer-Kellner, Martin Vechev

ICML 2022 · Label-efficient and robust learning › OOD and anomaly detection · approach: Uncertainty and misclassification detection · 4 citations (OpenAlex)

This paper tried a common change: Add a contrastive loss

Add contrastive learning · wrong-binop real-test · classification precision: 47.96 -> 52.30 Improved

What 1,317 accepted papers report for this change

See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.

Read the paper

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
Train first on synthetic bugs, then on imbalanced real bugsCuBERT bug detectors · var-misuse real-testclassification precision: 3.43 -> 64.79Improved
Train first on synthetic bugs, then on imbalanced real bugsCuBERT bug detectors · wrong-binop real-testclassification recall: 49.08 -> 43.99Mixed
Use real bugs rather than synthetic bugs in phase two (ablation)Two-phase CuBERT training · var-misuse real-testclassification precision: 35.59 -> 64.79Improved
Use real bugs rather than synthetic bugs in phase two (ablation)Two-phase CuBERT training · arg-swap real-testclassification precision: 44.19 -> 73.68Improved
Add a separate classification head (ablation)CuBERT bug detectors · var-misuse real-testclassification precision: 49.45 -> 64.79Improved
Add task hierarchy (ablation)CuBERT bug detectors · var-misuse real-testclassification precision: 51.82 -> 64.79Improved
Add focal loss (ablation)CuBERT bug detectors · wrong-binop real-testclassification precision: 49.32 -> 52.30Improved
Add contrastive learning (ablation)CuBERT bug detectors · wrong-binop real-testclassification precision: 47.96 -> 52.30Improved
Add focal loss (ablation)CuBERT bug detectors · arg-swap real-testclassification precision: 73.68 -> 73.68No clear effect
Increase the non-buggy-to-buggy ratio in real-data training (ablation)Two-phase CuBERT detector · var-misuse real-testaverage precision: 7.40 -> 20.96Improved
Increase the non-buggy-to-buggy ratio in real-data training (ablation)Two-phase CuBERT detector · wrong-binop real-testaverage precision: 19.84 -> 39.85Improved
Increase the non-buggy-to-buggy ratio in real-data training (ablation)Two-phase CuBERT detector · arg-swap real-testaverage precisionNo clear effect

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