arXiv subjects

Yinghua Li

Publications and source records attributed to Yinghua Li.

At least 19 recordsLinked to original sources

arXiv2026

Semantic Drift in Bug Resolution: How Behavioral Signals Propagate from Reports to Tests and Patches

Desc2Fix is a framework for measuring semantic alignment between bug reports, triggering tests, and developer-written fixes. Alignment is operationalized through structured behavioral anchors (e.g., reproduction steps, API/exception cues, expected vs. actual behavior), deterministic similarity metrics (ROUGE, SBERT, CodeBERT, OpenAI embeddings), and LLM-based judgments grounded in coverage, correctness, and specificity. Our analysis covers 2,857 report-test-patch triplets from Defects4J and SWT-Bench using two widely adopted instruction-tuned LLMs from distinct model families. LLMs reliably extract structured signals (up to 90% completeness) and exhibit strong cross-model consistency, yielding a stable semantic input contract for downstream reasoning. However, alignment is highly representation-sensitive: lexical similarity alone is insufficient, full diffs provide the most stable basis for judging report-patch correspondence, and structured summaries trade surface overlap for stronger correspondence at the level of individual actions and entities. Across more than 182,000 LLM-based alignment judgments, both models exhibit systematic optimism relative to humans (1-2 points on 5-point scales) and only modest rank agreement, motivating bias-aware evaluation. Behavioral alignment is measurable but not reducible to similarity, and structured anchors combined with embedding-based proxies provide reproducible signals for ranking and filtering tests and patches. Desc2Fix could support more reliable test generation, fault localization, patch ranking, and bug report authoring.

cs.SE↗

arXiv2026

Global strong solutions for 1D compressible Navier-Stokes/Cahn-Hilliard equations with vacuum

In this paper, we study the initial-boundary value problem of the 1D compressible Navier--Stokes/Cahn--Hilliard system with vacuum. We establish the global existence and uniqueness of strong solutions to this initial-boundary value problem. No any initial compatibility conditions are required via time weighted techniques, which leads to a loss of regularity near the initial time. Therefore, the uniqueness of solutions obtained in this paper is even more challenging. To address this issue, we establish refined growth estimates and singular-in-time weighted energy estimates that induce a Gronwall-type structure, which ultimately allows us to close the uniqueness proof in Eulerian coordinates without passing to Lagrangian coordinates.

math.AP↗

arXiv2026

Humanizing Automatically Generated Unit Test Suites with LLM-Based Refactoring

Search-based test generation tools such as EvoSuite produce compilable and high-coverage unit tests at scale, but their suites are often hard to read and maintain. LLMs can generate more natural tests, yet direct generation remains brittle, with compilation rates of only 51-78% in our study. We introduce TestHumanizer, a hybrid SBST+LLM approach that uses LLMs as controlled refactoring layers over compilable SBST suites to improve naming, structure, and developer-oriented clarity while preserving behavior and compilation validity. We evaluate TestHumanizer on 350 classes from Defects4J and SF110. EvoSuite generates 15 suites per class, and each suite is refactored under three context configurations using gpt-4o and mistral-large-2407, yielding 31,500 refactorings. TestHumanizer reaches 88-98% compilation rates, close to EvoSuite's 100% baseline and clearly above direct LLM generation. Structural coverage is largely preserved, typically within 1-2 percentage points, and 86-95% of refactorings satisfy a composite faithful-refactoring threshold. Refactored suites also improve predicted readability, reduce control-flow and cognitive complexity, and mitigate structural smells. The summary-based setting offers the most robust trade-off, while long code-centric prompts are more prone to hallucination-induced failures. A developer study on 30 classes and 444 test methods confirms significant gains in perceived readability and willingness to adopt, with Wilcoxon p less than 0.01 and substantial inter-rater agreement. Overall, LLMs are most effective not as standalone generators but as validation-gated refinement layers over robust SBST outputs.

cs.SE↗

arXiv2026

Well-posedness and blow-up criterion for strong solutions of the compressible Navier-Stokes/Allen-Cahn system with vacuum

This paper is devoted to the study of strong solutions for the compressible Navier-Stokes/Allen-Cahn system in bounded domain $Ω\subset\mathbb R^3$, allowing for the presence of initial vacuum. A characteristic of this system is the strong coupling between density and the Allen-Cahn equation, which leads to strong degeneracy in vacuum regions. Under a compatibility condition on the initial phase-field variable, we establish the local existence and uniqueness of strong solutions for $0\leρ_0\in W^{1,q}$ with $q\in(3,6)$, $u_0\in H_0^1$ and $χ_0\in H^2$. Owing to time-weighted estimates, no compatibility condition is required for the velocity, but these estimates introduce a singularity in proving uniqueness. We then establish a criterion for the possible breakdown of such a local strong solution at finite time in terms of blow-up of the quantities $\|\nabla u\|_{L_t^{1} L_x^{\infty}}$, $\|u\|_{L_t^{2} L_x^{\infty}}$ and $\|\nabla χ\|_{L_t^{2} L_x^{\infty}}$.

math.AP↗

arXiv2026

CARE: A Molecular-Guided Foundation Model with Adaptive Region Modeling for Whole Slide Image Analysis

Foundation models have recently achieved impressive success in computational pathology, demonstrating strong generalization across diverse histopathology tasks. However, existing models overlook the heterogeneous and non-uniform organization of pathological regions of interest (ROIs) because they rely on natural image backbones not tailored for tissue morphology. Consequently, they often fail to capture the coherent tissue architecture beyond isolated patches, limiting interpretability and clinical relevance. To address these challenges, we present Cross-modal Adaptive Region Encoder (CARE), a foundation model for pathology that automatically partitions WSIs into several morphologically relevant regions. Specifically, CARE employs a two-stage pretraining strategy: (1) a self-supervised unimodal pretraining stage that learns morphological representations from 34,277 whole-slide images (WSIs) without segmentation annotations, and (2) a cross-modal alignment stage that leverages RNA and protein profiles to refine the construction and representation of adaptive regions. This molecular guidance enables CARE to identify biologically relevant patterns and generate irregular yet coherent tissue regions, selecting the most representative area as ROI. CARE supports a broad range of pathology-related tasks, using either the ROI feature or the slide-level feature obtained by aggregating adaptive regions. Based on only one-tenth of the pretraining data typically used by mainstream foundation models, CARE achieves superior average performance across 33 downstream benchmarks, including morphological classification, molecular prediction, and survival analysis, and outperforms other foundation model baselines overall.

cs.CV↗

arXiv2026

Large-scale, Independent and Comprehensive study of the power of LLMs for test case generation

Unit testing is essential for software reliability, yet manual test creation is time-consuming and often neglected. Search-based software testing improves efficiency but produces tests with poor readability and maintainability, while LLMs show promise but lack comprehensive evaluation across reasoning-based prompting and real-world scenarios. This study presents the first large-scale empirical evaluation of LLM-generated unit tests at the full class level, analyzing four models (GPT-3.5, GPT-4, Mistral 7B, and Mixtral 8x7B) against EvoSuite across 216,300 test cases targeting Defects4J, SF110, and CMD. We evaluate five prompting techniques, ZSL, FSL, CoT, ToT, and GToT, assessing compilability, hallucination-driven failures, readability, coverage, and test smells. Reasoning-based prompting, particularly GToT, significantly enhances reliability and compilability, yet hallucination-driven failures remain persistent, with compilation failure rates reaching 86%. While LLM-generated tests are generally more readable than SBST outputs, recurring issues such as Magic Number Tests and Assertion Roulette hinder maintainability. These findings suggest that hybrid approaches combining LLM-based generation with automated validation and search-based refinement are necessary for production-ready results.

cs.SE↗

arXiv2025

Beyond Surface Similarity: Evaluating LLM-Based Test Refactorings with Structural and Semantic Awareness

Large Language Models (LLMs) are increasingly used to refactor unit tests, improving readability and structure while preserving behavior. Evaluating such refactorings, however, remains difficult: metrics like CodeBLEU penalize beneficial renamings and edits, while semantic similarities overlook readability and modularity. We propose CTSES, a first step toward human-aligned evaluation of refactored tests. CTSES combines CodeBLEU, METEOR, and ROUGE-L into a composite score that balances semantics, lexical clarity, and structural alignment. Evaluated on 5,000+ refactorings from Defects4J and SF110 (GPT-4o and Mistral-Large), CTSES reduces false negatives and provides more interpretable signals than individual metrics. Our emerging results illustrate that CTSES offers a proof-of-concept for composite approaches, showing their promise in bridging automated metrics and developer judgments.

cs.SE↗

arXiv2025

Human-Aligned Code Readability Assessment with Large Language Models

Code readability is crucial for software comprehension and maintenance, yet difficult to assess at scale. Traditional static metrics often fail to capture the subjective, context-sensitive nature of human judgments. Large Language Models (LLMs) offer a scalable alternative, but their behavior as readability evaluators remains underexplored. We introduce CoReEval, the first large-scale benchmark for evaluating LLM-based code readability assessment, comprising over 1.4 million model-snippet-prompt evaluations across 10 state of the art LLMs. The benchmark spans 3 programming languages (Java, Python, CUDA), 2 code types (functional code and unit tests), 4 prompting strategies (ZSL, FSL, CoT, ToT), 9 decoding settings, and developer-guided prompts tailored to junior and senior personas. We compare LLM outputs against human annotations and a validated static model, analyzing numerical alignment (MAE, Pearson's, Spearman's) and justification quality (sentiment, aspect coverage, semantic clustering). Our findings show that developer-guided prompting grounded in human-defined readability dimensions improves alignment in structured contexts, enhances explanation quality, and enables lightweight personalization through persona framing. However, increased score variability highlights trade-offs between alignment, stability, and interpretability. CoReEval provides a robust foundation for prompt engineering, model alignment studies, and human in the loop evaluation, with applications in education, onboarding, and CI/CD pipelines where LLMs can serve as explainable, adaptable reviewers.

cs.SE↗

arXiv2025

Rethinking Cognitive Complexity for Unit Tests: Toward a Readability-Aware Metric Grounded in Developer Perception

Automatically generated unit tests-from search-based tools like EvoSuite or LLMs-vary significantly in structure and readability. Yet most evaluations rely on metrics like Cyclomatic Complexity and Cognitive Complexity, designed for functional code rather than test code. Recent studies have shown that SonarSource's Cognitive Complexity metric assigns near-zero scores to LLM-generated tests, yet its behavior on EvoSuite-generated tests and its applicability to test-specific code structures remain unexplored. We introduce CCTR, a Test-Aware Cognitive Complexity metric tailored for unit tests. CCTR integrates structural and semantic features like assertion density, annotation roles, and test composition patterns-dimensions ignored by traditional complexity models but critical for understanding test code. We evaluate 15,750 test suites generated by EvoSuite, GPT-4o, and Mistral Large-1024 across 350 classes from Defects4J and SF110. Results show CCTR effectively discriminates between structured and fragmented test suites, producing interpretable scores that better reflect developer-perceived effort. By bridging structural analysis and test readability, CCTR provides a foundation for more reliable evaluation and improvement of generated tests. We publicly release all data, prompts, and evaluation scripts to support replication.

cs.SE↗

arXiv2025

Navier-Stokes/Allen-Cahn system with moving contact line

In this paper, we study a diffuse interface model for two-phase immiscible flows coupled by Navier-Stokes equations and mass-conserving Allen-Cahn equations. The contact line (the intersection of the fluid-fluid interface with the solid wall) moves along the wall when one fluid replaces the other, such as in liquid spreading or oil-water displacement. The system is equipped with the generalized Navier boundary conditions (GNBC) for the fluid velocity ${\boldsymbol u}$, and dynamic boundary condition or relaxation boundary condition for the phase field variable $ϕ$. We first obtain the local-in-time existence of unique strong solutions to the 2D and 3D Navier-Stokes/Allen-Cahn (NSAC) system with generalized Navier boundary conditions and dynamic boundary condition. For the 2D case in channels, we further show these solutions can be extended to any large time $T$. Additionally, we prove the local-in-time strong solutions for systems with generalized Navier boundary conditions and relaxation boundary condition in 3D channels. Finally, we establish a global unique strong solution accompany with some exponential decay estimates when the fluids are near phase separation states and the contact angle closes to 90 degrees or the fluid-fluid interface tension constant is small.

math.AP↗

arXiv2025

Well-posedness of the nonhomogeneous incompressible Navier-Stokes/Allen-Cahn system

In this paper, we investigate a system coupled by nonhomogeneous incompressible Navier-Stokes equations and Allen-Cahn equations describing a diffuse interface for two-phase flow of viscous fluids with different densities in a bounded domain $Ω\subset\mathbb R^d (d=2, 3)$. The mobility is allowed to depend on phase variable but non-degenerate. We first prove the existence of global weak solutions to the initial boundary value problem in 2D and 3D cases. Then we obtain the existence of local in time strong solutions in 3D case as well as the global strong solutions in 2D case. Moreover, by imposing smallness conditions on the initial data, the 3D local in time strong solution is extended globally, with an exponential decay rate for perturbations. At last, we show the weak-strong uniqueness.

math.AP↗

arXiv2025

Incompressible Limit of Strong Solutions to the Diffuse Interface Model for Two-phase Flows

This paper is concerned with the incompressible limit problem for strong solutions of compressible two-phase flow models under periodic boundary conditions, where the Navier-Stokes equations are nonlinearly coupled with either Cahn-Hilliard equations or Allen-Cahn equations. The viscosity coefficients are allowed to depend both on the density and the phase field variable. We establish rigorous convergence of both local and global strong solutions of compressible systems to their incompressible systems as the Mach number tends to zero.This theoretical framework establishes an essential linkage between compressible and incompressible phase field models, demonstrating that both formulations exhibit consistent physical fidelity in capturing interfacial flow dynamics.Furthermore, we provide some convergence rate estimates of the solutions.

math.AP↗

arXiv2024

Well-posedness of Navier-Stokes/Cahn-Hilliard equations modeling the dynamics of contact line in a channel

In this paper, we study the contact line problem in a channel. Precisely, we consider the incompressible Navier-Stokes/Cahn-Hilliard system with eneralized Navier boundary condition and relaxation boundary condition in a channel, which is the phase field model for the moving contact line problem in fluid mechanics. We establish the existence and uniqueness of local-in-time strong solution to this initial boundary value problem in 2D. To our knowledge, this is the first result to give the local-in-time well-posedness of Navier-Stokes/Cahn-Hilliard system with generalized Navier boundary condition and relaxation boundary condition. This result provides a rigorous mathematical analysis to confirm that the physical and numerical results by Qian-Wang-Sheng [Phys. Rev. E 68 (2003), 016306, 1-15; J. Fluid Mech. 564 (2006), 333-360] are well-posed and reasonable.

math.AP↗

arXiv2024

On the Diffusion of Test Smells in LLM-Generated Unit Tests

LLMs promise to transform unit test generation from a manual burden into an automated solution. Yet, beyond metrics such as compilability or coverage, little is known about the quality of LLM-generated tests, particularly their susceptibility to test smells, design flaws that undermine readability and maintainability. This paper presents the first multi-benchmark, large-scale analysis of test smell diffusion in LLM-generated unit tests. We contrast LLM outputs with human-written suites (as the reference for real-world practices) and SBST-generated tests from EvoSuite (as the automated baseline), disentangling whether LLMs reproduce human-like flaws or artifacts of synthetic generation. Our study draws on 20,505 class-level suites from four LLMs (GPT-3.5, GPT-4, Mistral 7B, Mixtral 8x7B), 972 method-level cases from TestBench, 14,469 EvoSuite tests, and 779,585 human-written tests from 34,635 open-source Java projects. Using two complementary detection tools (TsDetect and JNose), we analyze prevalence, co-occurrence, and correlations with software attributes and generation parameters. Results show that LLM-generated tests consistently manifest smells such as Assertion Roulette and Magic Number Test, with patterns strongly influenced by prompting strategy, context length, and model scale. Comparisons reveal overlaps with human-written tests, raising concerns of potential data leakage from training corpora while EvoSuite exhibits distinct, generator-specific flaws. These findings highlight both the promise and the risks of LLM-based test generation, and call for the design of smell-aware generation frameworks, prompt engineering strategies, and enhanced detection tools to ensure maintainable, high-quality test code.

cs.SE↗

arXiv2024

An Empirical Study of AI Techniques in Mobile Applications

The integration of artificial intelligence (AI) into mobile applications has significantly transformed various domains, enhancing user experiences and providing personalized services through advanced machine learning (ML) and deep learning (DL) technologies. AI-driven mobile apps typically refer to applications that leverage ML/DL technologies to perform key tasks such as image recognition and natural language processing. In this paper, we conducted the most extensive empirical study on AI applications, exploring on-device ML apps, on-device DL apps, and AI service-supported (cloud-based) apps. Our study encompasses 56,682 real-world AI applications, focusing on three crucial perspectives: 1) Application analysis, where we analyze the popularity of AI apps and investigate the update states of AI apps; 2) Framework and model analysis, where we analyze AI framework usage and AI model protection; 3) User analysis, where we examine user privacy protection and user review attitudes. Our study has strong implications for AI app developers, users, and AI R\&D. On one hand, our findings highlight the growing trend of AI integration in mobile applications, demonstrating the widespread adoption of various AI frameworks and models. On the other hand, our findings emphasize the need for robust model protection to enhance app security. Additionally, our study highlights the importance of user privacy and presents user attitudes towards the AI technologies utilized in current AI apps. We provide our AI app dataset (currently the most extensive AI app dataset) as an open-source resource for future research on AI technologies utilized in mobile applications.

cs.SE↗

arXiv2022

The Best of Both Worlds: Combining Learned Embeddings with Engineered Features for Accurate Prediction of Correct Patches

A large body of the literature on automated program repair develops approaches where patches are automatically generated to be validated against an oracle (e.g., a test suite). Because such an oracle can be imperfect, the generated patches, although validated by the oracle, may actually be incorrect. Our empirical work investigates different representation learning approaches for code changes to derive embeddings that are amenable to similarity computations of patch correctness identification, and assess the possibility of accurate classification of correct patch by combining learned embeddings with engineered features. Experimental results demonstrate the potential of learned embeddings to empower Leopard (a patch correctness predicting framework implemented in this work) with learning algorithms in reasoning about patch correctness: a machine learning predictor with BERT transformer-based learned embeddings associated with XGBoost achieves an AUC value of about 0.803 in the prediction of patch correctness on a new dataset of 2,147 labeled patches that we collected for the experiments. Our investigations show that deep learned embeddings can lead to complementary/better performance when comparing against the state-of-the-art, PATCH-SIM, which relies on dynamic information. By combining deep learned embeddings and engineered features, Panther (the upgraded version of Leopard implemented in this work) outperforms Leopard with higher scores in terms of AUC, +Recall and -Recall, and can accurately identify more (in)correct patches that cannot be predicted by the classifiers only with learned embeddings or engineered features. Finally, we use an explainable ML technique, SHAP, to empirically interpret how the learned embeddings and engineered features are contributed to the patch correctness prediction.

cs.SE↗

arXiv2022

Predicting Patch Correctness Based on the Similarity of Failing Test Cases

Towards predicting patch correctness in APR, we propose a simple, but novel hypothesis on how the link between the patch behaviour and failing test specifications can be drawn: similar failing test cases should require similar patches. We then propose BATS, an unsupervised learning-based system to predict patch correctness by checking patch Behaviour Against failing Test Specification. BATS exploits deep representation learning models for code and patches: for a given failing test case, the yielded embedding is used to compute similarity metrics in the search for historical similar test cases in order to identify the associated applied patches, which are then used as a proxy for assessing generated patch correctness. Experimentally, we first validate our hypothesis by assessing whether ground-truth developer patches cluster together in the same way that their associated failing test cases are clustered. Then, after collecting a large dataset of 1278 plausible patches (written by developers or generated by some 32 APR tools), we use BATS to predict correctness: BATS achieves an AUC between 0.557 to 0.718 and a recall between 0.562 and 0.854 in identifying correct patches. Compared against previous work, we demonstrate that our approach outperforms state-of-the-art performance in patch correctness prediction, without the need for large labeled patch datasets in contrast with prior machine learning-based approaches. While BATS is constrained by the availability of similar test cases, we show that it can still be complementary to existing approaches: used in conjunction with a recent approach implementing supervised learning, BATS improves the overall recall in detecting correct patches. We finally show that BATS can be complementary to the state-of-the-art PATCH-SIM dynamic approach of identifying the correct patches for APR tools.

cs.SE↗

arXiv2021

Strong Solutions for 1D Compressible Navier-Stokes/Allen-Cahn System with Phase Variable Dependent Viscosity

This paper is concerned with a non-isentropic compressible Navier-Stokes/Allen-Cahn system with phase variable dependent viscosity $η(χ)=χ^α$ and temperature dependent heat-conductivity $κ(θ)=θ^β$. We show the global existence of strong solutions under some assumptions on growth exponent $α$ and initial data. It is worth noting that the initial data could be large if $α\ge0$ is small, and the growth exponent $β>0$ can be arbitrary large.

math.AP↗