arXiv subjects

Yuekun Wang

Publications and source records attributed to Yuekun Wang.

4 recordsLinked to original sources

arXiv2026

PRWeaver: Evaluating LLM-Based Code Auditors against Long-Horizon Malicious Pull Requests

LLM-based code auditors are increasingly integrated into pull-request (PR) workflows, yet their reliability against adversarial changes distributed across repository evolution remains poorly understood. We introduce PRWeaver, a benchmark of 208 execution-validated attacks from ten real-world repositories, each instantiated under four matched review renderings (832 renderings in total). We evaluate three PR-auditing agents across six auditor-model systems. Across all systems, decomposing an attack changes detection by at most five percentage points, showing that commit boundaries alone do not explain evasion. In contrast, per-PR interleaving at $N=16$ and coherent carrier fusion reduce detection by 5-13 and 10-18 points, respectively. Under whole-window review at $N=24$, detection falls to 16-22%, compared with 50-60% under per-PR review. These results show that access to repository history is insufficient: concealment becomes most effective when benign and malicious changes jointly occupy the auditor's active review context or when the stated purpose plausibly accounts for the attack-bearing diff.

cs.SE↗

arXiv2025

Enhancing Automated Program Repair via Faulty Token Localization and Quality-Aware Patch Refinement

Large language models (LLMs) have recently demonstrated strong potential for automated program repair (APR). However, existing LLM-based techniques primarily rely on coarse-grained external feedback (e.g.,test results) to guide iterative patch generation, while lacking fine-grained internal signals that reveal why a patch fails or which parts of the generated code are likely incorrect. This limitation often leads to inefficient refinement, error propagation, and suboptimal repair performance. In this work, we propose TokenRepair, a novel two-level refinement framework that enhances APR by integrating internal reflection for localizing potentially faulty tokens with external feedback for quality-aware patch refinement. Specifically, TokenRepair first performs internal reflection by analyzing context-aware token-level uncertainty fluctuations to identify suspicious or low-confidence tokens within a patch. It then applies Chain-of-Thought guided rewriting to refine only these localized tokens, enabling targeted and fine-grained correction. To further stabilize the iterative repair loop, TokenRepair incorporates a quality-aware external feedback mechanism that evaluates patch quality and filters out low-quality candidates before refinement. Experimental results show that TokenRepair achieves new state-of-the-art repair performance, correctly fixing 88 bugs on Defects4J 1.2 and 139 bugs on HumanEval-Java, demonstrating substantial improvements ranging from 8.2% to 34.9% across all models on Defects4J 1.2 and from 3.3% to 16.1% on HumanEval-Java.

cs.SE↗

arXiv2025

Scheduzz: Constraint-based Fuzz Driver Generation with Dual Scheduling

Fuzzing a library requires experts to understand the library usage well and craft high-quality fuzz drivers, which is tricky and tedious. Therefore, many techniques have been proposed to automatically generate fuzz drivers. However, they fail to generate rational fuzz drivers due to the lack of adherence to proper library usage conventions, such as ensuring a resource is closed after being opened. To make things worse, existing library fuzzing techniques unconditionally execute each driver, resulting in numerous irrational drivers that waste computational resources while contributing little coverage and generating false positive bug reports. To tackle these challenges, we propose a novel automatic library fuzzing technique, Scheduzz, an LLM-based library fuzzing technique. It leverages LLMs to understand rational usage of libraries and extract API combination constraints. To optimize computational resource utilization, a dual scheduling framework is implemented to efficiently manage API combinations and fuzz drivers. The framework models driver generation and the corresponding fuzzing campaign as an online optimization problem. Within the scheduling loop, multiple API combinations are selected to generate fuzz drivers, while simultaneously, various optimized fuzz drivers are scheduled for execution or suspension. We implemented Scheduzz and evaluated it in 33 real-world libraries. Compared to baseline approaches, Scheduzz significantly reduces computational overhead and outperforms UTopia on 16 out of 21 libraries. It achieves 1.62x, 1.50x, and 1.89x higher overall coverage than the state-of-the-art techniques CKGFuzzer, Promptfuzz, and the handcrafted project OSS-Fuzz, respectively. In addition, Scheduzz discovered 33 previously unknown bugs in these well-tested libraries, 3 of which have been assigned CVEs.

cs.SE↗

arXiv2024

TGMM: Combining Parse Tree with GPU for Scalable Multilingual and Multi-Granularity Code Clone Detection

The rapid evolution of programming languages and software systems has necessitated the implementation of multilingual and scalable clone detection tools. However, it is difficult to achieve the above requirements at the same time. Most existing tools only focus on one challenge. In this work, we propose TGMM, a tree and GPU-based tool for multilingual and multi-granularity code clone detection. By generating parse trees based on user-provided grammar files, TGMM can extract code blocks at a specified granularity and detect Type-3 clones efficiently. In order to show the performance of TGMM, we compare it with seven state-of-the-art tools in terms of recall, precision, and execution time. TGMM ranks first in execution time and precision, while its recall is comparable to the others. Moreover, we analyzed the language extensibility of TGMM across 30 mainstream programming languages. Out of these, a total of 25 languages were supported, while the remaining five currently lack the necessary grammar files. Finally, we analyzed the clone characteristics of nine popular languages at five common granularities, hoping to inspire future researchers. The source code of TGMM is available at: https://github.com/TGMM24/TGMM.git.

cs.SE↗