Actionable Warning Is Not Enough: Recommending Valid Actionable Warnings with Weak Supervision

2025-11-15

AI Summary

Static analysis tools suffer from high false-positive rates, hindering their practical adoption; existing actionable warning identification methods rely on inaccurate assumptions, yielding numerous invalid recommendations. To address this, we propose ACWRecommender—the first weakly supervised learning framework for actionable warning recommendation. It constructs a large-scale, real-world dataset of actionable warnings and introduces a weak-labeling mechanism to model warning veracity. The framework adopts a two-stage paradigm: coarse-grained filtering using the UniXcoder pre-trained model, followed by fine-grained re-ranking to precisely identify and prioritize high-confidence true defects. Empirical evaluation across multiple projects demonstrates that the top-27 warnings recommended by ACWRecommender are all manually verified as genuine defects (out of 2,197 total warnings). It significantly outperforms baselines in nDCG and MRR, validating both its effectiveness and practical utility.

Technology Category

• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

The use of static analysis tools has gained increasing popularity among developers in the last few years. However, the widespread adoption of static analysis tools is hindered by their high false alarm rates. Previous studies have introduced the concept of actionable warnings and built a machine-learning method to distinguish actionable warnings from false alarms. However, according to our empirical observation, the current assumption used for actionable warning(s) collection is rather shaky and inaccurate, leading to a large number of invalid actionable warnings. To address this problem, in this study, we build the first large actionable warning dataset by mining 68,274 reversions from Top-500 GitHub C repositories, we then take one step further by assigning each actionable warning a weak label regarding its likelihood of being a real bug. Following that, we propose a two-stage framework called ACWRecommender to automatically recommend the actionable warnings with high probability to be real bugs (AWHB). Our approach warms up the pre-trained model UniXcoder by identifying actionable warnings task (coarse-grained detection stage) and rerank AWHB to the top by weakly supervised learning (fine-grained reranking stage). Experimental results show that our proposed model outperforms several baselines by a large margin in terms of nDCG and MRR for AWHB recommendation. Moreover, we ran our tool on 6 randomly selected projects and manually checked the top-ranked warnings from 2,197 reported warnings, we reported top-10 recommended warnings to developers, 27 of them were already confirmed by developers as real bugs. Developers can quickly find real bugs among the massive amount of reported warnings, which verifies the practical usage of our tool.

Problem

• Improving static analysis tools by reducing false alarm rates
• Addressing inaccurate actionable warnings through weak supervision
• Recommending high-probability real bugs from massive warning reports

Innovation

• Weak supervision labels actionable warning likelihood
• Two-stage framework warms up pre-trained model
• Weakly supervised reranking improves real bug detection