arXiv subjects

Cuiyun Gao

Publications and source records attributed to Cuiyun Gao.

At least 73 records · Page 4Linked to original sources

arXiv2025

MLLM-Based UI2Code Automation Guided by UI Layout Information

Converting user interfaces into code (UI2Code) is a crucial step in website development, which is time-consuming and labor-intensive. The automation of UI2Code is essential to streamline this task, beneficial for improving the development efficiency. There exist deep learning-based methods for the task; however, they heavily rely on a large amount of labeled training data and struggle with generalizing to real-world, unseen web page designs. The advent of Multimodal Large Language Models (MLLMs) presents potential for alleviating the issue, but they are difficult to comprehend the complex layouts in UIs and generate the accurate code with layout preserved. To address these issues, we propose LayoutCoder, a novel MLLM-based framework generating UI code from real-world webpage images, which includes three key modules: (1) Element Relation Construction, which aims at capturing UI layout by identifying and grouping components with similar structures; (2) UI Layout Parsing, which aims at generating UI layout trees for guiding the subsequent code generation process; and (3) Layout-Guided Code Fusion, which aims at producing the accurate code with layout preserved. For evaluation, we build a new benchmark dataset which involves 350 real-world websites named Snap2Code, divided into seen and unseen parts for mitigating the data leakage issue, besides the popular dataset Design2Code. Extensive evaluation shows the superior performance of LayoutCoder over the state-of-the-art approaches. Compared with the best-performing baseline, LayoutCoder improves 10.14% in the BLEU score and 3.95% in the CLIP score on average across all datasets.

cs.SE↗

arXiv2025

Boosting Vulnerability Detection of LLMs via Curriculum Preference Optimization with Synthetic Reasoning Data

Large language models (LLMs) demonstrate considerable proficiency in numerous coding-related tasks; however, their capabilities in detecting software vulnerabilities remain limited. This limitation primarily stems from two factors: (1) the absence of reasoning data related to vulnerabilities, which hinders the models' ability to capture underlying vulnerability patterns; and (2) their focus on learning semantic representations rather than the reason behind them, thus failing to recognize semantically similar vulnerability samples. Furthermore, the development of LLMs specialized in vulnerability detection is challenging, particularly in environments characterized by the scarcity of high-quality datasets. In this paper, we propose a novel framework ReVD that excels at mining vulnerability patterns through reasoning data synthesizing and vulnerability-specific preference optimization. Specifically, we construct forward and backward reasoning processes for vulnerability and corresponding fixed code, ensuring the synthesis of high-quality reasoning data. Moreover, we design the triplet supervised fine-tuning followed by curriculum online preference optimization for enabling ReVD to better understand vulnerability patterns. The extensive experiments conducted on PrimeVul and SVEN datasets demonstrate that ReVD sets new state-of-the-art for LLM-based software vulnerability detection, e.g., 12.24\%-22.77\% improvement in the accuracy. The source code and data are available at https://github.com/Xin-Cheng-Wen/PO4Vul.

cs.AI↗

arXiv2025

A Closer Look into Transformer-Based Code Intelligence Through Code Transformation: Challenges and Opportunities

Transformer-based models have demonstrated state-of-the-art performance in many intelligent coding tasks such as code comment generation and code completion. Previous studies show that deep learning models are sensitive to the input variations, but few studies have systematically studied the robustness of Transformer under perturbed input code. In this work, we empirically study the effect of semantic-preserving code transformation on the performance of Transformer. Specifically, 24 and 27 code transformation strategies are implemented for two popular programming languages, Java and Python, respectively. For facilitating analysis, the strategies are grouped into five categories: block transformation, insertion/deletion transformation, grammatical statement transformation, grammatical token transformation, and identifier transformation. Experiments on three popular code intelligence tasks, including code completion, code summarization and code search, demonstrate insertion/deletion transformation and identifier transformation show the greatest impact on the performance of Transformer. Our results also suggest that Transformer based on abstract syntax trees (ASTs) shows more robust performance than the model based on only code sequence under most code transformations. Besides, the design of positional encoding can impact the robustness of Transformer under code transformation. Based on our findings, we distill some insights about the challenges and opportunities for Transformer-based code intelligence.

cs.SE↗

arXiv2025

Exploring the Landscape of Text-to-SQL with Large Language Models: Progresses, Challenges and Opportunities

Converting natural language (NL) questions into SQL queries, referred to as Text-to-SQL, has emerged as a pivotal technology for facilitating access to relational databases, especially for users without SQL knowledge. Recent progress in large language models (LLMs) has markedly propelled the field of natural language processing (NLP), opening new avenues to improve text-to-SQL systems. This study presents a systematic review of LLM-based text-to-SQL, focusing on four key aspects: (1) an analysis of the research trends in LLM-based text-to-SQL; (2) an in-depth analysis of existing LLM-based text-to-SQL techniques from diverse perspectives; (3) summarization of existing text-to-SQL datasets and evaluation metrics; and (4) discussion on potential obstacles and avenues for future exploration in this domain. This survey seeks to furnish researchers with an in-depth understanding of LLM-based text-to-SQL, sparking new innovations and advancements in this field.

cs.CL↗

arXiv2025

RAG or Fine-tuning? A Comparative Study on LCMs-based Code Completion in Industry

Code completion, a crucial practice in industrial settings, helps developers improve programming efficiency by automatically suggesting code snippets during development. With the emergence of Large Code Models (LCMs), this field has witnessed significant advancements. Due to the natural differences between open-source and industrial codebases, such as coding patterns and unique internal dependencies, it is a common practice for developers to conduct domain adaptation when adopting LCMs in industry. There exist multiple adaptation approaches, among which retrieval-augmented generation (RAG) and fine-tuning are the two most popular paradigms. However, no prior research has explored the trade-off of the two approaches in industrial scenarios. To mitigate the gap, we comprehensively compare the two paradigms including Retrieval-Augmented Generation (RAG) and Fine-tuning (FT), for industrial code completion in this paper. In collaboration with Tencent's WXG department, we collect over 160,000 internal C++ files as our codebase. We then compare the two types of adaptation approaches from three dimensions that are concerned by industrial practitioners, including effectiveness, efficiency, and parameter sensitivity, using six LCMs. Our findings reveal that RAG, when implemented with appropriate embedding models that map code snippets into dense vector representations, can achieve higher accuracy than fine-tuning alone. Specifically, BM25 presents superior retrieval effectiveness and efficiency among studied RAG methods. Moreover, RAG and fine-tuning are orthogonal and their combination leads to further improvement. We also observe that RAG demonstrates better scalability than FT, showing more sustained performance gains with larger scales of codebase.

cs.SE↗

arXiv2025

LFTF: Locating First and Then Fine-Tuning for Mitigating Gender Bias in Large Language Models

Nowadays, Large Language Models (LLMs) have attracted widespread attention due to their powerful performance. However, due to the unavoidable exposure to socially biased data during training, LLMs tend to exhibit social biases, particularly gender bias. To better explore and quantifying the degree of gender bias in LLMs, we propose a pair of datasets named GenBiasEval and GenHintEval, respectively. The GenBiasEval is responsible for evaluating the degree of gender bias in LLMs, accompanied by an evaluation metric named AFGB-Score (Absolutely Fair Gender Bias Score). Meanwhile, the GenHintEval is used to assess whether LLMs can provide responses consistent with prompts that contain gender hints, along with the accompanying evaluation metric UB-Score (UnBias Score). Besides, in order to mitigate gender bias in LLMs more effectively, we present the LFTF (Locating First and Then Fine-Tuning) algorithm.The algorithm first ranks specific LLM blocks by their relevance to gender bias in descending order using a metric called BMI (Block Mitigating Importance Score). Based on this ranking, the block most strongly associated with gender bias is then fine-tuned using a carefully designed loss function. Numerous experiments have shown that our proposed LFTF algorithm can significantly mitigate gender bias in LLMs while maintaining their general capabilities.

cs.CL↗

arXiv2025

Smaller but Better: Self-Paced Knowledge Distillation for Lightweight yet Effective LCMs

Large code models (LCMs) have remarkably advanced the field of code generation. Despite their impressive capabilities, they still face practical deployment issues, such as high inference costs, limited accessibility of proprietary LCMs, and adaptability issues of ultra-large LCMs. These issues highlight the critical need for more accessible, lightweight yet effective LCMs. Knowledge distillation (KD) offers a promising solution, which transfers the programming capabilities of larger, advanced LCMs to smaller, less powerful LCMs. In this paper, we propose a novel Self-Paced knOwledge DistillAtion framework, named SODA, aiming at developing lightweight yet effective student LCMs. SODA consists of three stages in one cycle: (1) Correct-and-Fault Knowledge Delivery stage aims at improving the student models capability to recognize errors while ensuring its basic programming skill during the knowledge transferring, which involves correctness-aware supervised learning and fault-aware contrastive learning methods. (2) Multi-View Feedback stage aims at measuring the quality of results generated by the student model from two views, including model-based and static tool-based measurement, for identifying the difficult questions. (3) Feedback-based Knowledge Update stage aims at updating the student model adaptively by generating new questions at different difficulty levels, in which the difficulty levels are categorized based on the feedback in the second stage. Experimental results show that SODA improves the student model by 65.96% in terms of average Pass@1, outperforming the best baseline by 29.85%. Based on the SODA framework, we develop SodaCoder, a series of lightweight yet effective LCMs, which outperform 15 LCMs with less than or equal to 16B parameters. Notably, SodaCoder-DS-6.7B, built on DeepseekCoder-6.7B, even surpasses the prominent ChatGPT on average Pass@1.

cs.SE↗

arXiv2025

Towards Mitigating API Hallucination in Code Generated by LLMs with Hierarchical Dependency Aware

Application Programming Interfaces (APIs) are crucial in modern software development. Large Language Models (LLMs) assist in automated code generation but often struggle with API hallucination, including invoking non-existent APIs and misusing existing ones in practical development scenarios. Existing studies resort to Retrieval-Augmented Generation (RAG) methods for mitigating the hallucination issue, but tend to fail since they generally ignore the structural dependencies in practical projects and do not indeed validate whether the generated APIs are available or not. To address these limitations, we propose MARIN, a framework for mitigating API hallucination in code generated by LLMs with hierarchical dependency aware. MARIN consists of two phases: Hierarchical Dependency Mining, which analyzes local and global dependencies of the current function, aiming to supplement comprehensive project context in LLMs input, and Dependency Constrained Decoding, which utilizes mined dependencies to adaptively constrain the generation process, aiming to ensure the generated APIs align with the projects specifications. To facilitate the evaluation of the degree of API hallucination, we introduce a new benchmark APIHulBench and two new metrics including Micro Hallucination Number (MiHN) and Macro Hallucination Rate (MaHR). Experiments on six state-of-the-art LLMs demonstrate that MARIN effectively reduces API hallucinations, achieving an average decrease of 67.52% in MiHN and 73.56% in MaHR compared to the RAG approach. Applied to Huaweis internal projects and two proprietary LLMs, MARIN achieves average decreases of 57.33% in MiHN and 59.41% in MaHR.

cs.SE↗

arXiv2025

Can LLMs Replace Human Evaluators? An Empirical Study of LLM-as-a-Judge in Software Engineering

Recently, large language models (LLMs) have been deployed to tackle various software engineering (SE) tasks like code generation, significantly advancing the automation of SE tasks. However, assessing the quality of these LLM-generated code and text remains challenging. The commonly used Pass@k metric necessitates extensive unit tests and configured environments, demands a high labor cost, and is not suitable for evaluating LLM-generated text. Conventional metrics like BLEU, which measure only lexical rather than semantic similarity, have also come under scrutiny. In response, a new trend has emerged to employ LLMs for automated evaluation, known as LLM-as-a-judge. These LLM-as-a-judge methods are claimed to better mimic human assessment than conventional metrics without relying on high-quality reference answers. Nevertheless, their exact human alignment in SE tasks remains unexplored. In this paper, we empirically explore LLM-as-a-judge methods for evaluating SE tasks, focusing on their alignment with human judgments. We select seven LLM-as-a-judge methods that utilize general-purpose LLMs, alongside two LLMs specifically fine-tuned for evaluation. After generating and manually scoring LLM responses on three recent SE datasets of code translation, code generation, and code summarization, we then prompt these methods to evaluate each response. Finally, we compare the scores generated by these methods with human evaluation. The results indicate that output-based methods reach the highest Pearson correlation of 81.32 and 68.51 with human scores in code translation and generation, achieving near-human evaluation, noticeably outperforming ChrF++, one of the best conventional metrics, at 34.23 and 64.92. Such output-based methods prompt LLMs to output judgments directly, and exhibit more balanced score distributions that resemble human score patterns. Finally, we provide...

cs.SE↗

arXiv2025

SliceLocator: Locating Vulnerable Statements with Graph-based Detectors

Vulnerability detection is a crucial component in the software development lifecycle. Existing vulnerability detectors, especially those based on deep learning (DL) models, have achieved high effectiveness. Despite their capability of detecting vulnerable code snippets from given code fragments, the detectors are typically unable to further locate the fine-grained information pertaining to the vulnerability, such as the precise vulnerability triggering locations. Although explanation methods can filter important statements based on the predictions of code fragments, their effectiveness is limited by the fact that the model primarily learns the difference between vulnerable and non-vulnerable samples. In this paper, we propose SliceLocator, which, unlike previous approaches, leverages the detector's understanding of the differences between vulnerable and non-vulnerable samples, essentially, vulnerability-fixing statements. SliceLocator identifies the most relevant taint flow by selecting the highest-weighted flow path from all potential vulnerability-triggering statements in the program, in conjunction with the detector. We demonstrate that SliceLocator consistently performs well on four state-of-the-art GNN-based vulnerability detectors, achieving an accuracy of around 87% in flagging vulnerability-triggering statements across six common C/C++ vulnerabilities. It outperforms five widely used GNN-based explanation methods and two statement-level detectors.

cs.SE↗

arXiv2025

An Empirical Study of Retrieval-Augmented Code Generation: Challenges and Opportunities

Code generation aims to automatically generate code snippets of specific programming language according to natural language descriptions. The continuous advancements in deep learning, particularly pre-trained models, have empowered the code generation task to achieve remarkable performance. One main challenge of pre-trained models for code generation is the semantic gap between natural language requirements and source code. To address the issue, prior studies typically adopt a retrieval-augmented framework for the task, where the similar code snippets collected by a retrieval process can be leveraged to help understand the requirements and provide guidance for the generation process. However, there is a lack of systematic study on the application of this framework for code generation, including the impact of the final generated results and the specific usage of the framework. In this paper, we choose three popular pre-trained code models, namely CodeGen, UniXcoder, and CodeT5, to assess the impact of the quality and utilization of retrieved code on the retrieval-augmented framework. Our analysis shows that the retrieval-augmented framework is beneficial for improving the performance of the existing pre-trained models. We also provide suggestions on the utilization of the retrieval-augmented code generation framework: BM25 and Sequential Integration Fusion are recommended due to their convenience and superior performance. Sketch Filling Fusion, which extracts a sketch of relevant code, could help the model improve its performance further. Additionally, we conduct experiments to investigate the influence of the retrieval-augmented framework on large language models for code generation, showing the effectiveness of the framework, and we discuss the trade-off between performance improvement and computational costs in each phase within the framework.

cs.SE↗

arXiv2025

The Prompt Alchemist: Automated LLM-Tailored Prompt Optimization for Test Case Generation

Test cases are essential for validating the reliability and quality of software applications. Recent studies have demonstrated the capability of Large Language Models (LLMs) to generate useful test cases for given source code. However, the existing work primarily relies on human-written plain prompts, which often leads to suboptimal results since the performance of LLMs can be highly influenced by the prompts. Moreover, these approaches use the same prompt for all LLMs, overlooking the fact that different LLMs might be best suited to different prompts. Given the wide variety of possible prompt formulations, automatically discovering the optimal prompt for each LLM presents a significant challenge. Although there are methods on automated prompt optimization in the natural language processing field, they are hard to produce effective prompts for the test case generation task. First, the methods iteratively optimize prompts by simply combining and mutating existing ones without proper guidance, resulting in prompts that lack diversity and tend to repeat the same errors in the generated test cases. Second, the prompts are generally lack of domain contextual knowledge, limiting LLMs' performance in the task.

cs.SE↗

arXiv2024

The Current Challenges of Software Engineering in the Era of Large Language Models

With the advent of large language models (LLMs) in the artificial intelligence (AI) area, the field of software engineering (SE) has also witnessed a paradigm shift. These models, by leveraging the power of deep learning and massive amounts of data, have demonstrated an unprecedented capacity to understand, generate, and operate programming languages. They can assist developers in completing a broad spectrum of software development activities, encompassing software design, automated programming, and maintenance, which potentially reduces huge human efforts. Integrating LLMs within the SE landscape (LLM4SE) has become a burgeoning trend, necessitating exploring this emergent landscape's challenges and opportunities. The paper aims at revisiting the software development life cycle (SDLC) under LLMs, and highlighting challenges and opportunities of the new paradigm. The paper first summarizes the overall process of LLM4SE, and then elaborates on the current challenges based on a through discussion. The discussion was held among more than 20 participants from academia and industry, specializing in fields such as software engineering and artificial intelligence. Specifically, we achieve 26 key challenges from seven aspects, including software requirement & design, coding assistance, testing code generation, code review, code maintenance, software vulnerability management, and data, training, and evaluation. We hope the achieved challenges would benefit future research in the LLM4SE field.

cs.SE↗

arXiv2024

CodeRepoQA: A Large-scale Benchmark for Software Engineering Question Answering

In this work, we introduce CodeRepoQA, a large-scale benchmark specifically designed for evaluating repository-level question-answering capabilities in the field of software engineering. CodeRepoQA encompasses five programming languages and covers a wide range of scenarios, enabling comprehensive evaluation of language models. To construct this dataset, we crawl data from 30 well-known repositories in GitHub, the largest platform for hosting and collaborating on code, and carefully filter raw data. In total, CodeRepoQA is a multi-turn question-answering benchmark with 585,687 entries, covering a diverse array of software engineering scenarios, with an average of 6.62 dialogue turns per entry. We evaluate ten popular large language models on our dataset and provide in-depth analysis. We find that LLMs still have limitations in question-answering capabilities in the field of software engineering, and medium-length contexts are more conducive to LLMs' performance. The entire benchmark is publicly available at https://github.com/kinesiatricssxilm14/CodeRepoQA.

cs.SE↗

arXiv2024

Repository-Level Graph Representation Learning for Enhanced Security Patch Detection

Software vendors often silently release security patches without providing sufficient advisories (e.g., Common Vulnerabilities and Exposures) or delayed updates via resources (e.g., National Vulnerability Database). Therefore, it has become crucial to detect these security patches to ensure secure software maintenance. However, existing methods face the following challenges: (1) They primarily focus on the information within the patches themselves, overlooking the complex dependencies in the repository. (2) Security patches typically involve multiple functions and files, increasing the difficulty in well learning the representations. To alleviate the above challenges, this paper proposes a Repository-level Security Patch Detection framework named RepoSPD, which comprises three key components: 1) a repository-level graph construction, RepoCPG, which represents software patches by merging pre-patch and post-patch source code at the repository level; 2) a structure-aware patch representation, which fuses the graph and sequence branch and aims at comprehending the relationship among multiple code changes; 3) progressive learning, which facilitates the model in balancing semantic and structural information. To evaluate RepoSPD, we employ two widely-used datasets in security patch detection: SPI-DB and PatchDB. We further extend these datasets to the repository level, incorporating a total of 20,238 and 28,781 versions of repository in C/C++ programming languages, respectively, denoted as SPI-DB* and PatchDB*. We compare RepoSPD with six existing security patch detection methods and five static tools. Our experimental results demonstrate that RepoSPD outperforms the state-of-the-art baseline, with improvements of 11.90%, and 3.10% in terms of accuracy on the two datasets, respectively.

cs.SE↗

arXiv2024

EvalSVA: Multi-Agent Evaluators for Next-Gen Software Vulnerability Assessment

Software Vulnerability (SV) assessment is a crucial process of determining different aspects of SVs (e.g., attack vectors and scope) for developers to effectively prioritize efforts in vulnerability mitigation. It presents a challenging and laborious process due to the complexity of SVs and the scarcity of labeled data. To mitigate the above challenges, we introduce EvalSVA, a multi-agent evaluators team to autonomously deliberate and evaluate various aspects of SV assessment. Specifically, we propose a multi-agent-based framework to simulate vulnerability assessment strategies in real-world scenarios, which employs multiple Large Language Models (LLMs) into an integrated group to enhance the effectiveness of SV assessment in the limited data. We also design diverse communication strategies to autonomously discuss and assess different aspects of SV. Furthermore, we construct a multi-lingual SV assessment dataset based on the new standard of CVSS, comprising 699, 888, and 1,310 vulnerability-related commits in C++, Python, and Java, respectively. Our experimental results demonstrate that EvalSVA averagely outperforms the 44.12\% accuracy and 43.29\% F1 for SV assessment compared with the previous methods. It shows that EvalSVA offers a human-like process and generates both reason and answer for SV assessment. EvalSVA can also aid human experts in SV assessment, which provides more explanation and details for SV assessment.

cs.SE↗

arXiv2024

Split and Merge: Aligning Position Biases in LLM-based Evaluators

Large language models (LLMs) have shown promise as automated evaluators for assessing the quality of answers generated by AI systems. However, these LLM-based evaluators exhibit position bias, or inconsistency, when used to evaluate candidate answers in pairwise comparisons, favoring either the first or second answer regardless of content. To address this limitation, we propose PORTIA, an alignment-based system designed to mimic human comparison strategies to calibrate position bias in a lightweight yet effective manner. Specifically, PORTIA splits the answers into multiple segments, aligns similar content across candidate answers, and then merges them back into a single prompt for evaluation by LLMs. We conducted extensive experiments with six diverse LLMs to evaluate 11,520 answer pairs. Our results show that PORTIA markedly enhances the consistency rates for all the models and comparison forms tested, achieving an average relative improvement of 47.46%. Remarkably, PORTIA enables less advanced GPT models to achieve 88% agreement with the state-of-the-art GPT-4 model at just 10% of the cost. Furthermore, it rectifies around 80% of the position bias instances within the GPT-4 model, elevating its consistency rate up to 98%. Subsequent human evaluations indicate that the PORTIA-enhanced GPT-3.5 model can even surpass the standalone GPT-4 in terms of alignment with human evaluators. These findings highlight PORTIA's ability to correct position bias, improve LLM consistency, and boost performance while keeping cost-efficiency. This represents a valuable step toward a more reliable and scalable use of LLMs for automated evaluations across diverse applications.

cs.CL↗

arXiv2024

AEGIS: An Agent-based Framework for General Bug Reproduction from Issue Descriptions

In software maintenance, bug reproduction is essential for effective fault localization and repair. Manually writing reproduction scripts is a time-consuming task with high requirements for developers. Hence, automation of bug reproduction has increasingly attracted attention from researchers and practitioners. However, the existing studies on bug reproduction are generally limited to specific bug types such as program crashes, and hard to be applied to general bug reproduction. In this paper, considering the superior performance of agent-based methods in code intelligence tasks, we focus on designing an agent-based framework for the task. Directly employing agents would lead to limited bug reproduction performance, due to entangled subtasks, lengthy retrieved context, and unregulated actions. To mitigate the challenges, we propose an Automated gEneral buG reproductIon Scripts generation framework, named AEGIS, which is the first agent-based framework for the task. AEGIS mainly contains two modules: (1) A concise context construction module, which aims to guide the code agent in extracting structured information from issue descriptions, identifying issue-related code with detailed explanations, and integrating these elements to construct the concise context; (2) A FSM-based multi-feedback optimization module to further regulate the behavior of the code agent within the finite state machine (FSM), ensuring a controlled and efficient script generation process based on multi-dimensional feedback. Extensive experiments on the public benchmark dataset show that AEGIS outperforms the state-of-the-art baseline by 23.0% in F->P metric. In addition, the bug reproduction scripts generated by AEGIS can improve the relative resolved rate of Agentless by 12.5%.

cs.SE↗