arXiv subjects

Shin Hwei Tan

Publications and source records attributed to Shin Hwei Tan.

At least 19 recordsLinked to original sources

arXiv2026

Independent Patch Verification for Coding Agents with a Bidirectional Reconstruct-and-Verify Framework

Autonomous coding agents powered by large language models can now generate code patches directly from bug reports, but a fundamental gap remains: once a patch is produced, no mechanism independently verifies whether it truly resolves the reported problem. Prior work has sought to address this through iterative self-refinement and inference-time scaling, but these approaches either review the patch under the same interpretation that produced it or broaden candidate generation without verifying individual patches, and neither provides an explicit verification signal for assessing patch correctness. We propose RETRACE, a training-free post-generation verification framework that derives such a signal through bidirectional reconstruction and reconciliation. When a coding agent generates a candidate patch for an issue, RETRACE performs forward reconstruction to build an explicit repair rationale from the issue and the agent's trajectory; backward reconstruction then independently infers, from the patch and its trajectory alone and without access to the original issue, a description of the problem the patch appears to address, and compares this reconstruction against the original issue to produce an alignment verdict; a reconciliation stage then checks the consistency between the forward rationale and the patch, diagnoses the source of any misalignment, and either submits the patch or produces targeted revision guidance. Evaluated on SWE-bench Verified with two backbones (GPT-5-mini and MiniMax-2.5), RETRACE lifts Pass@1 by 7.0% and 3.6% respectively on the mini-SWE-agent scaffold, and delivers comparable gains on OpenHands without modification. Ablation experiments show that both the forward and backward stages contribute to the overall improvement and that adding reconciliation yields further gains.

cs.SE↗

arXiv2026

Understanding the Energy Impact of Software Refactoring: A Workload-Aware Study of Controlled Examples and Real-World Commits

Refactoring improves software maintainability while preserving functional behavior, yet behavior preservation does not imply energy neutrality. Existing studies primarily examine isolated refactorings under fixed or simple workloads, leaving the effects of workload variation, real-world refactoring practices, explanatory factors, and energy regression identification insufficiently understood. We present the first large-scale empirical study of the energy impact of refactoring across two complementary Java benchmarks: a Micro-benchmark, comprising 68 refactoring types evaluated under diverse workloads, and a Practical-benchmark, containing 481 real-world refactoring commits from 430 GitHub projects. Using repeated paired energy measurements, we analyze workload sensitivity, refactoring patterns, explanatory factors, and the effectiveness of metric- and LLM-based regression identification. In the Micro-benchmark, 199 of 384 refactoring-workload pairs (51.8%) exhibit statistically significant energy differences, and 45.3% of refactoring instances change energy-impact classification across workloads. In the Practical-benchmark, only 36 commits (7.5%) show significant energy changes, although two-thirds differ by at least 10%. Refactoring type alone is insufficient to predict energy outcomes, while certain recurring refactoring combinations are associated with energy reductions. Changes in execution time consistently explain energy variation in the controlled benchmark but correlate weakly with energy changes in real-world commits. Our findings highlight the need for workload-diverse evaluation of the energy impact of refactoring; neither existing metric-based approaches nor LLM-based predictors can reliably identify refactoring-induced energy regressions, motivating the development of more accurate techniques for predicting the energy impact of refactoring.

cs.SE↗

arXiv2026

From Codebases to LLMs: Non-Inclusive Naming in Linux Foundation Repositories

Since 2020, the Linux Foundation and the multi-organization Inclusive Naming Initiative (INI) have encouraged open-source projects to replace non-inclusive terms such as master/slave and whitelist/blacklist. Although these recommendations have been widely adopted, there is limited empirical evidence on their long-term adoption across Linux Foundation (LF) projects or their implications for AI-assisted software development. In this paper, we present NISCAN, a multilingual static-analysis framework that detects non-inclusive terminology across source code and related software artifacts using the INI vocabulary. Using NISCAN, we conduct the first ecosystem-scale study of inclusive naming across 461 Linux Foundation repositories. Our analysis shows that non-inclusive terminology has declined by approximately 47% since 2020, yet adoption remains incomplete: 62.7% of repositories still contain at least one Tier-1 non-inclusive identifier, while most remaining terminology resides outside source code in documentation, comments, configuration files, and other software artifacts. We further show that repository size, programming language, project functionality, and ecosystem are stronger predictors of term inclusiveness in LF repositories rather than foundation governance. To examine the implications for AI-assisted software development, we conduct a case study evaluating whether large language models (LLMs) can reconstruct legacy non-inclusive identifiers from surrounding program context. The results show that historical naming decisions remain embedded in model predictions even after identifiers have been renamed. Overall, our study findings provide the first ecosystem-scale assessment of inclusive naming adoption within the Linux Foundation and highlight the importance of addressing terminology residue to support responsible naming and ethically sourced code generation.

cs.SE↗

arXiv2026

LLM-Guided Issue Generation from Uncovered Code Segments

Developers are increasingly overwhelmed by AI-generated issue reports that lack actionability and reproducibility, eroding trust in automated bug detection tools. In this paper, we present IssueSpecter, an automated tool that finds bugs in uncovered code segments and automatically generates prioritized, actionable issue reports. IssueSpecter combines coverage analysis with LLM-based defect identification, producing structured reports complete with severity ratings, reproduction steps, and suggested fixes. We evaluate IssueSpecter on 13 actively maintained Python projects, generating 10,467 issue reports. Manual annotation of the top-130 ranked issues by IssueSpecter confirms that 84.6% of the LLM-generated issues are valid or warrant further investigation, with only 15.4% false positives. LLM-based ranking outperforms rule-based ranking by 50% at P@3 and 41% in MRR. The identified bugs cover a wide variety of types, from logic and boundary errors to security vulnerabilities and state consistency bugs. By ranking issues by priority, IssueSpecter aims to help developers focus their attention on the most impactful bugs first. Finally, we validate IssueSpecter through case studies reproducing real bugs surfaced from its generated issue reports, demonstrating its practical value for automatic bug discovery in open-source Python projects. Compared against CoverUp, a state-of-the-art coverage-driven test generation tool, IssueSpecter achieves a higher bug validity rate (81.0% vs. 76.2%) under identical evaluation conditions, using the same model and the same number of evaluated artifacts per project, while additionally providing structured issue reports with reproduction steps and candidate fixes that are immediately actionable without requiring developers to interpret generated test intent.

cs.SE↗

arXiv2026

Ethics Testing: Proactive Identification of Generative AI System Harms

Generative Artificial Intelligence (GAI) systems that can automatically generate content in the form of source code or other contents (e.g., images) has seen increasing popularity due to the emergence of tools such as ChatGPT which rely on Large Language Models (LLMs). Misuse of the automatically generated content can incur serious consequences due to potential harms in the generated content. Despite the importance of ensuring the quality of automatically generated content, there is little to no approach that can systematically generate tests for identifying software harms in the content generated by these GAI systems. In this article, we introduce the novel concept of ethics testing which aims to systematically generate tests for identifying software harms. Different from existing testing methodologies (e.g., fairness testing that aims to identifying software discrimination), ethics testing aims to systematically detect software harms that could be induced due to unethical behavior (e.g., harmful behavior or behavior that violates intellectual property rights) in automatically generated content. We introduced the concept of ethics testing, discussed the challenges therewithin, and conducted five case studies to show how ethics testing can be performed for generative AI systems.

cs.SE↗

arXiv2026

Think Before You Code: Dual Reasoning for the NLSafety-Utility Trade-Off in LLM Code Generation

Large language models (LLMs) for code generation are typically evaluated on functional correctness alone, overlooking whether generated code propagates harmful content embedded in the prompt. Prior work has shown that most Code LLMs reproduce offensive identifiers from injected renaming instructions without warning, yet existing approaches focus on detecting harmful content, neglecting functional correctness. Grounded in the Theory of Dual Channel Constraints (which states that code is a dual-channel medium combining an algorithmic (AL) channel for machine execution and a natural language (NL) channel for human communication, creating a unique safety-utility trade-off where a model must balance functional execution with responsible communication), we propose NLSafety-Utility Duality Score (SUDS), a metric that unifies code utility, safety adherence, and warning awareness into a single score across 12 ranked response scenarios, and Dual Reasoning (DR), a structured inference-time technique that requires an explicit safety audit and task-grounded code review before code generation. Evaluated on six LLMs across two benchmarks augmented with harmful keyword injections (820 and 2,135 samples), DR consistently achieves the highest SUDS across all models, improving mean SUDS by 1.32$\times$ to 3.42$\times$ over the baseline, while chain-of-thought prompting yields negligible safety gains and a safety-aware prompt provides only partial improvement. Further analysis reveals that DR's effectiveness scales with model capacity, that the one-shot exemplar primarily stabilizes output format for smaller models, and that structured reasoning cannot compensate for models with limited safety vocabularies.

cs.SE↗

arXiv2026

Investigating Code Reuse in Software Redesign: A Case Study

Software redesign preserves functionality while improving quality attributes, but manual reuse of code and tests is costly and error-prone, especially in crossrepository redesigns. Focusing on static analyzers where cross-repo redesign needs often arise, we conduct a bidirectional study of the ongoing Soot/SootUp redesign case using an action research methodology that combines empirical investigation with validated open-source contributions. Our study reveals: (1) non-linear migration which necessitates bidirectional reuse, (2) deferred reuse via TODOs, (3) neglected test porting, and (4) residual bug propagation during migrations. We identify tracking corresponding code and tests as the key challenge, and address it by retrofitting clone detection to derive code mappings between original and redesigned projects. Guided by semantic reuse patterns derived in our study, we propose Semantic Alignment Heuristics and a scalable hierarchical detection strategy. Evaluations on two redesigned project pairs (Soot/SootUp and FindBugs/SpotBugs) show that our approach achieves an average reduction of 33-99% in likely irrelevant clones at a SAS threshold of 0.5 across all tool results, and improves precision up to 86% on our benchmark of 1,749 samples. Moreover, we contribute to the redesigned projects by submitting five issues and 10 pull requests, of which eight have been merged.

cs.SE↗

arXiv2026

Understanding Bugs in Modern Agentic Frameworks: A Study of Symptoms, Root Causes, and Triggering Conditions

Modern agentic frameworks such as CrewAI and AutoGen have evolved into complex, autonomous multi-agent systems, introducing reliability challenges that go beyond earlier pipeline-based LLM libraries. However, existing empirical studies focus on earlier LLM libraries or task-level bugs, leaving the unique complexities of these agentic frameworks unexplored. We present a comprehensive study of 409 fixed bugs across five representative agentic frameworks, proposing a five-layer architectural abstraction. Our taxonomy identifies previously unreported symptom categories---Unexpected Execution Sequence, User Configuration Ignored, and Incomplete/Incorrect Trace---and isolates agent-specific root causes including Model-Related Fault, Cognitive Context Mismanagement, and Orchestration Fault. Notably, the model integration layer is the most bug-prone yet receives disproportionately low test inclusion rate during bug fixing (47%), revealing a critical validation gap. Despite varying design paradigms, bug symptoms, root causes, and bug-prone components show substantial cross-framework consistency (JS similarity 0.62--0.88). Finally, we present the first systematic study of bug-triggering conditions, identifying error-prone factor combinations across element configurations, input patterns, and operations, and demonstrate their transferability across frameworks, providing a foundation for test oracle design and cross-framework benchmark.

cs.SE↗

arXiv2026

COBOLAssist: Analyzing and Fixing Compilation Errors for LLM-Powered COBOL Code Generation

Legacy programming languages such as COBOL (Common Business-Oriented Language) remain critical in business computing. However, maintaining legacy COBOL systems is increasingly challenging due to a declining pool of skilled developers and the persistence of COBOL errors that require deep domain expertise to resolve. This paper investigates the challenges of COBOL compilation errors and introduces a framework leveraging large language models (LLMs) to address these issues. We first categorize the common compilation errors in LLM-generated COBOL code into three groups: incomplete code errors, syntax errors, and type-related errors. We further propose COBOLAssist, a technique to enhance code correctness through iterative repairs guided by compilation feedback. Our evaluation using five LLMs including GPT variants and mAInframer, shows a high prevalence of incorrect program structures and function usage in COBOL programs and demonstrates the effectiveness of COBOLAssist, with the compilation success rates increasing from 29.5\% to 64.38\% for GPT-4o-mini and from 41.8\% to 95.89\% for GPT-4o. It also improves pass@1 significantly, for example from 9.1 to 22.6 for GPT-4. Notably, while mAInframer-34B achieves the highest compilation success rate, its functional correctness remains limited. This research not only highlights the limitations in current LLMs for COBOL but also demonstrates a practical path forward for automated debugging in legacy systems.

cs.SE↗

arXiv2026

COBOL-Coder: Domain-Adapted Large Language Models for COBOL Code Generation and Translation

COBOL remains a critical language for mainframe systems, yet existing large language models (LLMs) struggle to generate and translate COBOL code correctly. This paper reports our experience in developing and evaluating domain-adapted LLMs for COBOL and mainframe software engineering. We introduce (1) an automated data curation pipeline that combines compiler-guided validation with multi-stage similarity-based filtering to construct high-quality COBOL training data, and (2) COBOL-Coder, a COBOL-specialized LLM fine-tuned on the curated COBOL domain data. We evaluate COBOL-Coder on two tasks: code generation (on COBOLEval and COBOLCodeBench) and code translation (on COBOL-JavaTrans, our proposed benchmark for bidirectional COBOL-Java translation). In our experiments, COBOL-Coder achieves up to a 73.95 percent compilation success rate and 49.33 Pass-1 on COBOLEval, compared to 41.8 percent and 16.4 for GPT-4o, while most open-source baselines (e.g., CodeGemma, CodeLlama, StarCoder2) fail to produce compilable programs. For Java-to-COBOL translation, COBOL-Coder reaches 34.93 Pass-1, whereas general-purpose LLMs achieve near-zero scores. To assess the usability of LLM-generated code in real-world settings, we conduct a survey with experienced COBOL developers. Participants consistently report that COBOL-Coder exhibits stronger COBOL awareness, has more reliable program structure, and is better aligned with enterprise practices than general-purpose LLMs.

cs.SE↗

arXiv2026

Probe to Generate: Program Variant-Guided Test Augmentation for Repository-Level Repair Benchmarks

Test-based benchmarks such as SWE-bench have become a standard basis for evaluating automated issue resolution agents, deeming a patch correct if it passes a provided regression test suite. In practice, weak test suites can admit plausible but semantically incorrect patches, inflating reported agent performance. We present \tool, a test augmentation framework that uses semantically modified program variants as behavioral probes to identify and close gaps in benchmark test suites. Variants of the reference patch that survive the original tests reveal under-constrained behaviors, which then guide targeted regression test generation. Each generated test is retained only if it passes on the reference patch, fails on at least one surviving variant, and remains robust under behavior-preserving transformations. On SWE-bench Verified, 77% of instances admit at least one surviving variant. \tool generates 1,014 validated tests across 211 instances, increasing patch-region line and branch coverage by 10.8 and 9.5 percentage points. Re-evaluating the top-10 repair agents with the augmented suites reduces resolved rates by 4.2%-9.0%, showing that many previously accepted patches exploit benchmark test gaps rather than fully satisfying the intended repair semantics. These findings demonstrate that benchmark evaluation is not solely a patch-generation problem but also a test-strength problem.

cs.SE↗

arXiv2026

What Makes Code Generation Ethically Sourced?

Several code generation models have been proposed to help reduce time and effort in solving software-related tasks. To ensure responsible AI, there are growing interests over various ethical issues (e.g., unclear licensing, privacy, fairness, and environment impact). These studies have the overarching goal of ensuring ethically sourced generation, which has gained growing attentions in speech synthesis and image generation. In this paper, we introduce the novel notion of Ethically Sourced Code Generation (ES-CodeGen) to refer to managing all processes involved in code generation model development from data collection to post-deployment via ethical and sustainable practices. To build a taxonomy of ES-CodeGen, we perform a two-phase literature review where we read 803 papers across various domains and specific to AI-based code generation. We identified 71 relevant papers with 10 initial dimensions of ES-CodeGen. To refine our dimensions and gain insights on consequences of ES-CodeGen, we surveyed 32 practitioners, which include six developers who submitted GitHub issues to opt-out from the Stack dataset (these impacted users have real-world experience of ethically sourcing issues in code generation models). The results lead to 11 dimensions of ES-CodeGen with a new dimension on code quality as practitioners have noted its importance. We also identified consequences, artifacts, and stages relevant to ES-CodeGen. Our post-survey reflection showed that most practitioners tend to ignore social-related dimensions despite their importance. Most practitioners either agreed or strongly agreed that our survey help improve their understanding of ES-CodeGen. Our study calls for attentions of various ethical issues towards ES-CodeGen.

cs.SE↗

arXiv2026

From Historical Patches to Repair Plans: Outcome-Conditioned Reasoning for Repository-Level Program Repair

Repository-level automated program repair (APR) requires long-horizon reasoning over interdependent decisions. However, most LLM-based approaches reconstruct repair reasoning independently for each issue, failing to reuse successful patterns from prior repairs, even though real-world repositories contain many related issues with shared structure or constraints. Existing methods typically rely on forward exploration, which operates under outcome uncertainty, incurs substantial inference-time overhead, and can drift from the final correct patch. We propose Conditional Reasoning Distillation (ConRAD), which leverages in-repository resolved issues by reconstructing repair reasoning backward from verified patches and distilling outcome-consistent, stage-wise repair reasoning plans. Injected at inference time, these plans guide fault localization and patch generation, replacing open-ended exploration with constrained inference without fine-tuning or search. On SWE-Bench Lite, ConRAD improves Pass@1 by 10.4\% (GPT-4o), 8.6\% (DeepSeek-V3), and 10.3\% (GPT-5), demonstrating a scalable inference-time alternative to forward exploration for long-horizon APR.

cs.SE↗

arXiv2025

BloomAPR: A Bloom's Taxonomy-based Framework for Assessing the Capabilities of LLM-Powered APR Solutions

Recent advances in large language models (LLMs) have accelerated the development of AI-driven automated program repair (APR) solutions. However, these solutions are typically evaluated using static benchmarks such as Defects4J and SWE-bench, which suffer from two key limitations: (1) the risk of data contamination, potentially inflating evaluation results due to overlap with LLM training data, and (2) limited ability to assess the APR capabilities in dynamic and diverse contexts. In this paper, we introduced BloomAPR, a novel dynamic evaluation framework grounded in Bloom's Taxonomy. Our framework offers a structured approach to assess the cognitive capabilities of LLM-powered APR solutions across progressively complex reasoning levels. Using Defects4J as a case study, we evaluated two state-of-the-art LLM-powered APR solutions, ChatRepair and CigaR, under three different LLMs: GPT-3.5-Turbo, Llama-3.1, and StarCoder-2. Our findings show that while these solutions exhibit basic reasoning skills and effectively memorize bug-fixing patterns (fixing up to 81.57% of bugs at the Remember layer), their performance increases with synthetically generated bugs (up to 60.66% increase at the Understand layer). However, they perform worse on minor syntactic changes (fixing up to 43.32% at the Apply layer), and they struggle to repair similar bugs when injected into real-world projects (solving only 13.46% to 41.34% bugs at the Analyze layer). These results underscore the urgent need for evolving benchmarks and provide a foundation for more trustworthy evaluation of LLM-powered software engineering solutions.

cs.SE↗

arXiv2025

Automated Harmfulness Testing for Code Large Language Models

Generative AI systems powered by Large Language Models (LLMs) usually use content moderation to prevent harmful content spread. To evaluate the robustness of content moderation, several metamorphic testing techniques have been proposed to test content moderation software. However, these techniques mainly focus on general users (e.g., text and image generation). Meanwhile, a recent study shows that developers consider using harmful keywords when naming software artifacts to be an unethical behavior. Exposure to harmful content in software artifacts can negatively impact the mental health of developers, making content moderation for Code Large Language Models (Code LLMs) essential. We conduct a preliminary study on program transformations that can be misused to introduce harmful content into auto-generated code, identifying 32 such transformations. To address this, we propose CHT, a coverage-guided harmfulness testing framework that generates prompts using diverse transformations and harmful keywords injected into benign programs. CHT evaluates output damage to assess potential risks in LLM-generated explanations and code. Our evaluation of four Code LLMs and GPT-4o-mini reveals that content moderation in LLM-based code generation is easily bypassed. To enhance moderation, we propose a two-phase approach that first detects harmful content before generating output, improving moderation effectiveness by 483.76\%.

cs.SE↗

arXiv2025

Aligning the Objective of LLM-based Program Repair

Large language models (LLMs) have achieved decent results on automated program repair (APR). However, the next token prediction training objective of decoder-only LLMs (e.g., GPT-4) is misaligned with the masked span prediction objective of current infilling-style methods, which impedes LLMs from fully leveraging pre-trained knowledge for program repair. In addition, while some LLMs can locate and repair bugs in certain functions using the related artifacts (e.g., test cases), existing methods still depend on statement-level fault localization methods to provide a list of buggy hunks for repair. This restriction hinders LLMs from exploring potential patches beyond the given locations. In this paper, we investigate a new approach to adapt LLMs to program repair. Our core insight is that LLM's APR capability can be greatly improved by simply aligning the output to their training objective and allowing them to refine the whole program without first identifying faulty statements. Based on this insight, we designed D4C, a straightforward prompting framework for APR. D4C can repair 180 bugs correctly in Defects4J, with each patch being sampled only 10 times. This surpasses the SOTA APR methods with perfect fault localization by 10% and reduces the patch sampling number by 90%. Our findings reveal that (1) objective alignment is crucial for fully exploiting LLM's pre-trained capability, and (2) replacing the traditional localize-buggy-hunks-then-repair workflow with direct debugging is more effective for LLM-based APR methods. Thus, we believe this paper introduces a new mindset for harnessing LLMs in APR.

cs.SE↗

arXiv2025

Guiding ChatGPT to Fix Web UI Tests via Explanation-Consistency Checking

The rapid evolution of Web UI incurs time and effort in UI test maintenance. Prior techniques in Web UI test repair focus on locating the target elements on the new Webpage that match the old ones so that the corresponding broken statements can be repaired. These techniques usually rely on prioritizing certain attributes (e.g., XPath) during matching where the similarity of certain attributes is ranked before other attributes, indicating that there may be bias towards certain attributes during matching. To mitigate the bias, we present the first study that investigates the feasibility of using prior Web UI repair techniques for initial matching and then using ChatGPT to perform subsequent matching. Our key insight is that given a list of elements matched by prior techniques, ChatGPT can leverage language understanding to perform subsequent matching and use its code generation model for fixing the broken statements. To mitigate hallucination in ChatGPT, we design an explanation validator that checks if the provided explanation for the matching results is consistent, and provides hints to ChatGPT via a self-correction prompt to further improve its results. Our evaluation on a widely used dataset shows that the ChatGPT-enhanced techniques improve the effectiveness of existing Web test repair techniques. Our study also shares several important insights in improving future Web UI test repair techniques.

cs.SE↗

arXiv2025

Testing Refactoring Engine via Historical Bug Report driven LLM

Refactoring is the process of restructuring existing code without changing its external behavior while improving its internal structure. Refactoring engines are integral components of modern Integrated Development Environments (IDEs) and can automate or semi-automate this process to enhance code readability, reduce complexity, and improve the maintainability of software products. Similar to traditional software systems such as compilers, refactoring engines may also contain bugs that can lead to unexpected behaviors. In this paper, we propose a novel approach called RETESTER, a LLM-based framework for automated refactoring engine testing. Specifically, by using input program structure templates extracted from historical bug reports and input program characteristics that are error-prone, we design chain-of-thought (CoT) prompts to perform refactoring-preserving transformations. The generated variants are then tested on the latest version of refactoring engines using differential testing. We evaluate RETESTER on two most popular modern refactoring engines (i.e., ECLIPSE, and INTELLIJ IDEA). It successfully revealed 18 new bugs in the latest version of those refactoring engines. By the time we submit our paper, seven of them were confirmed by their developers, and three were fixed.

cs.SE↗