arXiv subjects

Gunel Jahangirova

Publications and source records attributed to Gunel Jahangirova.

At least 19 recordsLinked to original sources

arXiv2026

COMPAS: Difficulty-Aware Joint Search for Optimizing Code Generation

Code generation systems make each LLM call with a model, a prompt, and decoding settings. However, existing optimization methods usually tune only part of these choices or use one fixed configuration for all tasks: global optimizers search one configuration for all tasks, routers choose only a model, and prompt optimizers keep the model and decoding settings fixed. This leaves their joint, group-specific interactions unclear. We therefore examine how these choices interact and observe that prompts and decoding settings interact, tuning effects vary by model, and the best configuration varies by task difficulty. Guided by these observations, we introduce COMPAS (Code-generation Optimization over Models, Prompts, And Decoding Settings), a difficulty-aware method that learns group-specific quality-cost fronts through low-cost model selection and joint prompt-decoding search, then routes each test task to its matching front online without further search. Under a matched search budget on LiveCodeBench, COMPAS improves pass@1 from 45.9% for the best baseline to 52.8% while reducing cost from $36.57 to $4.92. This also transfers to repository-level code generation on SWE-bench, resolving 76.0% of tasks versus 70.0% for the best baseline. Code and the reproducibility artifact are available at https://github.com/gjz78910/COMPAS.

cs.SE↗

arXiv2026

An Empirical Study of Downstream Adaptation for Agent Skills

As Large Language Model (LLM) agents become integral to modern software systems, ``skills'' have emerged as a novel unit of software reuse, enabling developers to package workflows, decision procedures, and prompt-based policies. While skills are intended for reuse, downstream developers frequently modify published skills to fit local contexts, yet little is known about the nature of such adaptations. This paper presents the first empirical study of downstream skill adaptation in public forks, to understand how published skills are adapted, and to provide implications for researchers and engineers on improving skill design, evolution, and orchestration. Specifically, we analyze 1,126 skill-adaptation instances from six widely adopted skill repositories and develop a taxonomy comprising 46 adaptation patterns organized into 13 families. Our key findings reveal a reuse paradox: although skills are intended to be easily imported and reused, developers spend a lot of effort rewriting what the skills do, fixing skill discoverability, and translating them for different tools and languages, indicating a need for better abstractions, standardized interfaces, and automated support for skill adaptation. Furthermore, adaptations are highly interdependent, with changes in one component often requiring coordinated updates elsewhere, motivating automated support for detecting inconsistent modifications. We also find that nearly one-fifth of adaptations introduce security-sensitive content within the same instruction text that governs behavior.

cs.SE↗

arXiv2026

How Does Chunking Affect Retrieval-Augmented Code Completion? A Controlled Empirical Study

Retrieval-augmented generation (RAG) pipelines for code completion rely on chunking to segment source files into retrievable units, yet chunking strategies are typically adopted without empirical justification, and practitioner recommendations are notably inconsistent. We present a controlled empirical study isolating the effect of chunking on code completion quality by crossing four representative strategies (Function, Declaration, Sliding Window, and cAST) with four retrievers, five generators, and nine parameter configurations on two benchmarks (RepoEval and CrossCodeEval), totaling 864 experimental settings. Our results reveal that chunking strategy has a statistically significant effect on RAG-based code completion. Contrary to intuition, chunking based on functions underperforms all other strategies by 3.57--5.64 percentage points on RepoEval (Cliff's delta = -1.0), while the remaining chunking strategies perform comparably. Our further analysis demonstrates that this observation holds across all retriever--generator combinations. We also find that cross-file context length is the dominant parameter: doubling from 2,048 to 8,192 tokens yields up to 4.2 percentage points of improvement, whereas chunk size has a weaker, non-monotonic effect. On the cost--quality Pareto front, Sliding Window and cAST dominate both benchmarks; Function chunking is never Pareto-optimal.

cs.SE↗

arXiv2026

Benchmarking and Evaluating VLMs for Software Architecture Diagram Understanding

Software architecture diagrams are important design artifacts for communicating system structure, behavior, and data organization throughout the software development lifecycle. Although recent progress in large language models has substantially advanced code-centric software engineering tasks such as code generation, testing, and maintenance, the ability of modern vision-language models (VLMs) to understand software architecture diagrams remains underexplored. To address this gap, we present SADU, a benchmark for Software Architecture Diagram Understanding that evaluates VLMs on architecture diagrams as structured software engineering artifacts rather than generic images. SADU contains 154 carefully curated diagrams spanning behavioral, structural, and ER diagrams, paired with structured annotations and 2,431 question-answer tasks covering counting and retrieval reasoning. We evaluate 11 state-of-the-art VLMs from the Gemini, Claude, GPT, and Qwen families. Our results show that software architecture diagram understanding remains challenging for current models: the best-performing model gemini-3-flash-preview achieves only 70.18\% accuracy, while gpt-4o-mini only achieves 17.77\% accuracy. The results further reveal the weaknesses in diagram reasoning and visual relation grounding, highlighting a gap between current VLMs and the needs of design-stage software engineering. SADU provides a foundation for future research on diagram-aware AI systems and more faithful AI-assisted software engineering workflows.

cs.SE↗

arXiv2026

LLMLOOP: Improving LLM-Generated Code and Tests through Automated Iterative Feedback Loops

Large Language Models (LLMs) are showing remarkable performance in generating source code, yet the generated code often has issues like compilation errors or incorrect code. Researchers and developers often face wasted effort in implementing checks and refining LLM-generated code, frequently duplicating their efforts. This paper presents LLMLOOP, a framework that automates the refinement of both source code and test cases produced by LLMs. LLMLOOP employs five iterative loops: resolving compilation errors, addressing static analysis issues, fixing test case failures, and improving test quality through mutation analysis. These loops ensure the generation of high-quality test cases that serve as both a validation mechanism and a regression test suite for the generated code. We evaluated LLMLOOP on HUMANEVAL-X, a recent benchmark of programming tasks. Results demonstrate the tool's effectiveness in refining LLM-generated outputs.

cs.SE↗

arXiv2026

A Taxonomy of Real Faults in Hybrid Quantum-Classical Architectures

With the popularity of Hybrid Quantum-Classical architectures, particularly noisy intermediate-scale quantum (NISQ) architectures, comes the need for quality assurance methods tailored to their specific faults. In this study, we propose a taxonomy of faults in Hybrid Quantum-Classical architectures accompanied by a dataset of real faults in the identified categories. To achieve this, we empirically analysed open-source repositories for fixed faults. We analysed over 5000 closed issues on GitHub and pre-selected 529 of them based on rigorously defined inclusion criteria. We selected 133 faults that we labelled around symptoms and the origin of the faults. We cross-validated the classification and labels assigned to every fault between two of the authors. As a result, we introduced a taxonomy of real faults in Hybrid Quantum-Classical architectures. Subsequently, we validated the taxonomy through interviews conducted with eleven developers. The taxonomy was dynamically updated throughout the cross-validation and interview processes. The final version was validated and discussed through surveys conducted with an independent group of domain experts to ensure its relevance and to gain further insights.

cs.SE↗

arXiv2026

Revisiting "Revisiting Neuron Coverage for DNN Testing: A Layer-Wise and Distribution-Aware Criterion": A Critical Review and Implications on DNN Coverage Testing

We present a critical review of Neural Coverage (NLC), a state-of-the-art DNN coverage criterion by Yuan et al. at ICSE 2023. While NLC proposes to satisfy eight design requirements and demonstrates strong empirical performance, we question some of their theoretical and empirical assumptions. We observe that NLC deviates from core principles of coverage criteria, such as monotonicity and test suite order independence, and could more fully account for key properties of the covariance matrix. Additionally, we note threats to the validity of the empirical study, related to the ground truth ordering of test suites. Through our empirical validation, we substantiate our claims and propose improvements for future DNN coverage metrics. Finally, we conclude by discussing the implications of these insights.

cs.SE↗

arXiv2026

Understanding LLM-Driven Test Oracle Generation

Automated unit test generation aims to improve software quality while reducing the time and effort required for creating tests manually. However, existing techniques primarily generate regression oracles that predicate on the implemented behavior of the class under test. They do not address the oracle problem: the challenge of distinguishing correct from incorrect program behavior. With the rise of Foundation Models (FMs), particularly Large Language Models (LLMs), there is a new opportunity to generate test oracles that reflect intended behavior. This positions LLMs as enablers of Promptware, where software creation and testing are driven by natural-language prompts. This paper presents an empirical study on the effectiveness of LLMs in generating test oracles that expose software failures. We investigate how different prompting strategies and levels of contextual input impact the quality of LLM-generated oracles. Our findings offer insights into the strengths and limitations of LLM-based oracle generation in the FM era, improving our understanding of their capabilities and fostering future research in this area.

cs.SE↗

arXiv2025

New Formulation of DNN Statistical Mutation Killing for Ensuring Monotonicity: A Technical Report

Mutation testing has emerged as a powerful technique for evaluating the effectiveness of test suites for Deep Neural Networks. Among existing approaches, the statistical mutant killing criterion of DeepCrime has leveraged statistical testing to determine whether a mutant significantly differs from the original model. However, it suffers from a critical limitation: it violates the monotonicity property, meaning that expanding a test set may result in previously killed mutants no longer being classified as killed. In this technical report, we propose a new formulation of statistical mutant killing based on Fisher exact test that preserves the statistical rigour of it while ensuring monotonicity.

cs.SE↗

arXiv2025

Fault Localisation and Repair for DL Systems: An Empirical Study with LLMs

Numerous Fault Localisation (FL) and repair techniques have been proposed to address faults in Deep Learning (DL) models. However, their effectiveness in practical applications remains uncertain due to the reliance on pre-defined rules. This paper presents a comprehensive evaluation of state-of-the-art FL and repair techniques, examining their advantages and limitations. Moreover, we introduce a novel approach that harnesses the power of Large Language Models (LLMs) in localising and repairing DL faults. Our evaluation, conducted on a carefully designed benchmark, reveals the strengths and weaknesses of current FL and repair techniques. We emphasise the importance of enhanced accuracy and the need for more rigorous assessment methods that employ multiple ground truth patches. Notably, LLMs exhibit remarkable performance in both FL and repair tasks. For instance, the GPT-4 model achieves 44% and 82% improvements in FL and repair tasks respectively, compared to the second-best tool, demonstrating the potential of LLMs in this domain. Our study sheds light on the current state of FL and repair techniques and suggests that LLMs could be a promising avenue for future advancements.

cs.SE↗

arXiv2025

Comparative Analysis of Carbon Footprint in Manual vs. LLM-Assisted Code Development

Large Language Models (LLM) have significantly transformed various domains, including software development. These models assist programmers in generating code, potentially increasing productivity and efficiency. However, the environmental impact of utilising these AI models is substantial, given their high energy consumption during both training and inference stages. This research aims to compare the energy consumption of manual software development versus an LLM-assisted approach, using Codeforces as a simulation platform for software development. The goal is to quantify the environmental impact and propose strategies for minimising the carbon footprint of using LLM in software development. Our results show that the LLM-assisted code generation leads on average to 32.72 higher carbon footprint than the manual one. Moreover, there is a significant correlation between task complexity and the difference in the carbon footprint of the two approaches.

cs.SE↗

arXiv2025

MuFF: Stable and Sensitive Post-training Mutation Testing for Deep Learning

Rapid adoptions of Deep Learning (DL) in a broad range of fields led to the development of specialised testing techniques for DL systems, including DL mutation testing. However, existing post-training DL mutation techniques often generate unstable mutants across multiple training repetitions and multiple applications of the same mutation operator. Additionally, while extremely efficient, they generate mutants without taking into account the mutants' sensitivity and killability, resulting in a large number of ineffective mutants compared to pre-training mutants. In this paper, we present a new efficient post-training DL mutation technique, named MuFF, designed to ensure the stability of the mutants and capable of generating killable and sensitive mutants. MuFF implements an automated stability check and introduces two mutation operators, named weight and neuron inhibitors. Our extensive empirical experiments show that MuFF generates mutants with 60%pt and 25%pt higher sensitivity compared to DeepMutation++ and DeepCrime, respectively, while also producing mutants that are more stable than those of DeepMutation++ and different from the mutants of DeepCrime. Moreover, MuFF preserves the benefits of the post-training mutation technique, being 61 times faster than DeepCrime in generating mutants.

cs.SE↗

arXiv2024

Real Faults in Deep Learning Fault Benchmarks: How Real Are They?

As the adoption of Deep Learning (DL) systems continues to rise, an increasing number of approaches are being proposed to test these systems, localise faults within them, and repair those faults. The best attestation of effectiveness for such techniques is an evaluation that showcases their capability to detect, localise and fix real faults. To facilitate these evaluations, the research community has collected multiple benchmarks of real faults in DL systems. In this work, we perform a manual analysis of 490 faults from five different benchmarks and identify that 314 of them are eligible for our study. Our investigation focuses specifically on how well the bugs correspond to the sources they were extracted from, which fault types are represented, and whether the bugs are reproducible. Our findings indicate that only 18.5% of the faults satisfy our realism conditions. Our attempts to reproduce these faults were successful only in 52% of cases.

cs.SE↗

arXiv2024

An Empirical Study of Fault Localisation Techniques for Deep Learning

With the increased popularity of Deep Neural Networks (DNNs), increases also the need for tools to assist developers in the DNN implementation, testing and debugging process. Several approaches have been proposed that automatically analyse and localise potential faults in DNNs under test. In this work, we evaluate and compare existing state-of-the-art fault localisation techniques, which operate based on both dynamic and static analysis of the DNN. The evaluation is performed on a benchmark consisting of both real faults obtained from bug reporting platforms and faulty models produced by a mutation tool. Our findings indicate that the usage of a single, specific ground truth (e.g., the human defined one) for the evaluation of DNN fault localisation tools results in pretty low performance (maximum average recall of 0.31 and precision of 0.23). However, such figures increase when considering alternative, equivalent patches that exist for a given faulty DNN. Results indicate that \dfd is the most effective tool, achieving an average recall of 0.61 and precision of 0.41 on our benchmark.

cs.SE↗

arXiv2024

muPRL: A Mutation Testing Pipeline for Deep Reinforcement Learning based on Real Faults

Reinforcement Learning (RL) is increasingly adopted to train agents that can deal with complex sequential tasks, such as driving an autonomous vehicle or controlling a humanoid robot. Correspondingly, novel approaches are needed to ensure that RL agents have been tested adequately before going to production. Among them, mutation testing is quite promising, especially under the assumption that the injected faults (mutations) mimic the real ones. In this paper, we first describe a taxonomy of real RL faults obtained by repository mining. Then, we present the mutation operators derived from such real faults and implemented in the tool muPRL. Finally, we discuss the experimental results, showing that muPRL is effective at discriminating strong from weak test generators, hence providing useful feedback to developers about the adequacy of the generated test scenarios.

cs.SE↗

arXiv2024

GenMorph: Automatically Generating Metamorphic Relations via Genetic Programming

Metamorphic testing is a popular approach that aims to alleviate the oracle problem in software testing. At the core of this approach are Metamorphic Relations (MRs), specifying properties that hold among multiple test inputs and corresponding outputs. Deriving MRs is mostly a manual activity, since their automated generation is a challenging and largely unexplored problem. This paper presents GenMorph, a technique to automatically generate MRs for Java methods that involve inputs and outputs that are boolean, numerical, or ordered sequences. GenMorph uses an evolutionary algorithm to search for effective test oracles, i.e., oracles that trigger no false alarms and expose software faults in the method under test. The proposed search algorithm is guided by two fitness functions that measure the number of false alarms and the number of missed faults for the generated MRs. Our results show that GenMorph generates effective MRs for 18 out of 23 methods (mutation score >20%). Furthermore, it can increase Randoop's fault detection capability in 7 out of 23 methods, and Evosuite's in 14 out of 23 methods. When compared with AutoMR, a state-of-the-art MR generator, GenMorph also outperformed its fault detection capability in 9 out of 10 methods.

cs.SE↗

arXiv2023

Repairing DNN Architecture: Are We There Yet?

As Deep Neural Networks (DNNs) are rapidly being adopted within large software systems, software developers are increasingly required to design, train, and deploy such models into the systems they develop. Consequently, testing and improving the robustness of these models have received a lot of attention lately. However, relatively little effort has been made to address the difficulties developers experience when designing and training such models: if the evaluation of a model shows poor performance after the initial training, what should the developer change? We survey and evaluate existing state-of-the-art techniques that can be used to repair model performance, using a benchmark of both real-world mistakes developers made while designing DNN models and artificial faulty models generated by mutating the model code. The empirical evaluation shows that random baseline is comparable with or sometimes outperforms existing state-of-the-art techniques. However, for larger and more complicated models, all repair techniques fail to find fixes. Our findings call for further research to develop more sophisticated techniques for Deep Learning repair.

cs.SE↗

arXiv2021

DeepMetis: Augmenting a Deep Learning Test Set to Increase its Mutation Score

Deep Learning (DL) components are routinely integrated into software systems that need to perform complex tasks such as image or natural language processing. The adequacy of the test data used to test such systems can be assessed by their ability to expose artificially injected faults (mutations) that simulate real DL faults. In this paper, we describe an approach to automatically generate new test inputs that can be used to augment the existing test set so that its capability to detect DL mutations increases. Our tool DeepMetis implements a search based input generation strategy. To account for the non-determinism of the training and the mutation processes, our fitness function involves multiple instances of the DL model under test. Experimental results show that \tool is effective at augmenting the given test set, increasing its capability to detect mutants by 63% on average. A leave-one-out experiment shows that the augmented test set is capable of exposing unseen mutants, which simulate the occurrence of yet undetected faults.

cs.SE↗