arXiv subjects

Jongmoon Baik

Publications and source records attributed to Jongmoon Baik.

4 recordsLinked to original sources

arXiv2026

ReDef: Do Code Language Models Truly Understand Code Changes for Just-in-Time Software Defect Prediction?

Just-in-Time software defect prediction (JIT-SDP) plays a critical role in prioritizing risky code changes during code review and continuous integration. However, existing datasets often suffer from noisy labels and low precision in identifying bug-inducing commits. To address this, we present ReDef (Revert-based Defect dataset), a high-confidence benchmark of function-level modifications curated from 22 large-scale C/C++ projects. Defective cases are anchored by revert commits, while clean cases are validated through post-hoc history checks. Ambiguous instances are conservatively filtered out via a GPT-assisted triage process involving multiple votes and audits. This pipeline yields 3,164 defective and 10,268 clean modifications, offering substantially more reliable labels than prior resources. Beyond dataset construction, we provide a systematic evaluation of how Code Language Models (CLMs)-specifically CodeBERT, CodeT5+, UniXcoder, and Qwen2.5-reason about code modifications. We first investigate which input encodings most effectively expose change information under five different strategies. We then design four counterfactual perturbation strategies (e.g., swapping added/deleted blocks, inverting diff polarity) to serve as diagnostic probes. We posit that if models genuinely capture change semantics, such distortions should lead to a clear decline in predictive performance. Our results show that compact diff-style encodings consistently outperform whole-function formats across all CLMs, supported by rigorous statistical confirmation. However, under counterfactual tests, performance remains effectively stable, revealing that what appears to be robustness in fact reflects a reliance on superficial cues rather than true semantic understanding.

cs.SE↗

arXiv2025

Deep Synthetic Cross-Project Approaches for Software Reliability Growth Modeling

Software Reliability Growth Models (SRGMs) are widely used to predict software reliability based on defect discovery data collected during testing or operational phases. However, their predictive accuracy often degrades in data-scarce environments, such as early-stage testing or safety-critical systems. Although cross-project transfer learning has been explored to mitigate this issue by leveraging data from past projects, its applicability remains limited due to the scarcity and confidentiality of real-world datasets. To overcome these limitations, we propose Deep Synthetic Cross-project SRGM (DSC-SRGM), a novel approach that integrates synthetic data generation with cross-project transfer learning. Synthetic datasets are generated using traditional SRGMs to preserve the statistical characteristics of real-world defect discovery trends. A cross-correlation-based clustering method is applied to identify synthetic datasets with patterns similar to the target project. These datasets are then used to train a deep learning model for reliability prediction. The proposed method is evaluated on 60 real-world datasets, and its performance is compared with both traditional SRGMs and cross-project deep learning models trained on real-world datasets. DSC-SRGM achieves up to 23.3% improvement in predictive accuracy over traditional SRGMs and 32.2% over cross-project deep learning models trained on real-world datasets. However, excessive use of synthetic data or a naive combination of synthetic and real-world data may degrade prediction performance, highlighting the importance of maintaining an appropriate data balance. These findings indicate that DSC-SRGM is a promising approach for software reliability prediction in data-scarce environments.

cs.SE↗

arXiv2025

Retrieval-Augmented Code Review Comment Generation

Automated code review comment generation (RCG) aims to assist developers by automatically producing natural language feedback for code changes. Existing approaches are primarily either generation-based, using pretrained language models, or information retrieval-based (IR), reusing comments from similar past examples. While generation-based methods leverage code-specific pretraining on large code-natural language corpora to learn semantic relationships between code and natural language, they often struggle to generate low-frequency but semantically important tokens due to their probabilistic nature. In contrast, IR-based methods excel at recovering such rare tokens by copying from existing examples but lack flexibility in adapting to new code contexts-for example, when input code contains identifiers or structures not found in the retrieval database. To bridge the gap between generation-based and IR-based methods, this work proposes to leverage retrieval-augmented generation (RAG) for RCG by conditioning pretrained language models on retrieved code-review exemplars. By providing relevant examples that illustrate how similar code has been previously reviewed, the model is better guided to generate accurate review comments. Our evaluation on the Tufano et al. benchmark shows that RAG-based RCG outperforms both generation-based and IR-based RCG. It achieves up to +1.67% higher exact match and +4.25% higher BLEU scores compared to generation-based RCG. It also improves the generation of low-frequency ground-truth tokens by up to 24.01%. We additionally find that performance improves as the number of retrieved exemplars increases.

cs.SE↗

arXiv2025

Improving LLM-Based Fault Localization with External Memory and Project Context

Fault localization, the process of identifying the software components responsible for failures, is essential but often time-consuming. Recent advances in Large Language Models (LLMs) have enabled fault localization without extensive defect datasets or model fine-tuning. However, existing LLM-based methods rely only on general LLM capabilities and lack integration of project-specific knowledge, resulting in limited effectiveness, especially for complex software. We introduce MemFL, a novel approach that enhances LLM-based fault localization by integrating project-specific knowledge via external memory. This memory includes static summaries of the project and dynamic, iterative debugging insights gathered from previous attempts. By leveraging external memory, MemFL simplifies debugging into three streamlined steps, significantly improving efficiency and accuracy. Iterative refinement through dynamic memory further enhances reasoning quality over time. Evaluated on the Defects4J benchmark, MemFL using GPT-4o-mini localized 12.7% more bugs than current LLM-based methods, achieving this improvement with just 21% of the execution time (17.4 seconds per bug) and 33% of the API cost (0.0033 dollars per bug). On complex projects, MemFL's advantage increased to 27.6%. Additionally, MemFL with GPT-4.1-mini outperformed existing methods by 24.4%, requiring only 24.7 seconds and 0.0094 dollars per bug. MemFL thus demonstrates significant improvements by effectively incorporating project-specific knowledge into LLM-based fault localization, delivering high accuracy with reduced time and cost.

cs.SE↗