arXiv subjects

Jingzhu He

Publications and source records attributed to Jingzhu He.

5 recordsLinked to original sources

arXiv2026

SemiScope: Disentangling Classifier Tuning and Joint Optimization in Semi-Supervised Security Classification

Background. Labeled data for security classification is scarce. Semi-supervised learning (SSL) propagates labels from a small labeled pool to larger unlabeled pools. Yet security applications often use SSL as a black box: default parameters, a fixed classifier, and no handling of pseudo-label-induced class imbalance. Aims. Recent work reports sizeable gains from optimizing SSL pipelines via joint search, AutoML, or per-component tuning. These gains are hard to attribute: they may reflect useful SSL-classifier interactions, or mostly from simply tuning the downstream classifier. We disentangle these effects for binary tabular security data with classical SSL and tree-based classifiers. Method. We build SemiScope as an analysis instrument, not a deployment recommendation. It uses Bayesian Optimization to jointly tune SSL settings, confidence filtering, oversampling, and the classifier. The key control, Tuned-Clf, fixes SSL to defaults but gets the same 100-trial classifier budget and validation-set threshold tuning as SemiScope. At 10% labels, we compare them with paired TOST using a +/-1.0 g-measure smallest effect of interest. Results. SemiScope beats every default SSL baseline on all five datasets, improving over the strongest by 0.7-12.7 points. Under the equal-budget control, Tuned-Clf is statistically equivalent to the full pipeline on 4 of 5 datasets; Phishing is inconclusive. Classifier HPO alone recovers a median 86% of SemiScope's gain over Default Self-Training (ST) + Random Forest (RF). Conclusions. The reusable contribution is the decomposition protocol. A simpler recipe suffices: use Self-Training, tune the classifier with Bayesian Optimization, and tune the decision threshold on validation data. It reaches within 1 g-measure of Supervised RF at 20-30% labels on four datasets and 40% on Drebin, at the same or lower label rate than Default ST + RF on every dataset.

cs.LG↗

arXiv2025

Artemis: Toward Accurate Detection of Server-Side Request Forgeries through LLM-Assisted Inter-Procedural Path-Sensitive Taint Analysis

Server-side request forgery (SSRF) vulnerabilities are inevitable in PHP web applications. Existing static tools in detecting vulnerabilities in PHP web applications neither contain SSRF-related features to enhance detection accuracy nor consider PHP's dynamic type features. In this paper, we present Artemis, a static taint analysis tool for detecting SSRF vulnerabilities in PHP web applications. First, Artemis extracts both PHP built-in and third-party functions as candidate source and sink functions. Second, Artemis constructs both explicit and implicit call graphs to infer functions' relationships. Third, Artemis performs taint analysis based on a set of rules that prevent over-tainting and pauses when SSRF exploitation is impossible. Fourth, Artemis analyzes the compatibility of path conditions to prune false positives. We have implemented a prototype of Artemis and evaluated it on 250 PHP web applications. Artemis reports 207 true vulnerable paths (106 true SSRFs) with 15 false positives. Of the 106 detected SSRFs, 35 are newly found and reported to developers, with 24 confirmed and assigned CVE IDs.

cs.CR↗

arXiv2024

An LLM-based Readability Measurement for Unit Tests' Context-aware Inputs

Automated test techniques usually generate unit tests with higher code coverage than manual tests. However, the readability of automated tests is crucial for code comprehension and maintenance. The readability of unit tests involves many aspects. In this paper, we focus on test inputs. The central limitation of existing studies on input readability is that they focus on test codes alone without taking the tested source codes into consideration, making them either ignore different source codes' different readability requirements or require manual efforts to write readable inputs. However, we observe that the source codes specify the contexts that test inputs must satisfy. Based on such observation, we introduce the \underline{C}ontext \underline{C}onsistency \underline{C}riterion (a.k.a, C3), which is a readability measurement tool that leverages Large Language Models to extract primitive-type (including string-type) parameters' readability contexts from the source codes and checks whether test inputs are consistent with those contexts. We have also proposed EvoSuiteC3. It leverages C3's extracted contexts to help EvoSuite generate readable test inputs. We have evaluated C3's performance on $409$ \java{} classes and compared manual and automated tests' readability under C3 measurement. The results are two-fold. First, The Precision, Recall, and F1-Score of C3's mined readability contexts are \precision{}, \recall{}, and \fone{}, respectively. Second, under C3's measurement, the string-type input readability scores of EvoSuiteC3, ChatUniTest (an LLM-based test generation tool), manual tests, and two traditional tools (EvoSuite and Randoop) are $90\%$, $83\%$, $68\%$, $8\%$, and $8\%$, showing the traditional tools' inability in generating readable string-type inputs.

cs.SE↗

arXiv2024

Coverage Goal Selector for Combining Multiple Criteria in Search-Based Unit Test Generation

Unit testing is critical to the software development process, ensuring the correctness of basic programming units in a program (e.g., a method). Search-based software testing (SBST) is an automated approach to generating test cases. SBST generates test cases with genetic algorithms by specifying the coverage criterion (e.g., branch coverage). However, a good test suite must have different properties, which cannot be captured using an individual coverage criterion. Therefore, the state-of-the-art approach combines multiple criteria to generate test cases. Since combining multiple coverage criteria brings multiple objectives for optimization, it hurts the test suites' coverage for certain criteria compared with using the single criterion. To cope with this problem, we propose a novel approach named \textbf{smart selection}. Based on the coverage correlations among criteria and the subsumption relationships among coverage goals, smart selection selects a subset of coverage goals to reduce the number of optimization objectives and avoid missing any properties of all criteria. We conduct experiments to evaluate smart selection on $400$ Java classes with three state-of-the-art genetic algorithms under the $2$-minute budget. On average, smart selection outperforms combining all goals on $65.1\%$ of the classes having significant differences between the two approaches. Secondly, we conduct experiments to verify our assumptions about coverage criteria relationships. Furthermore, we assess the coverage performance of smart selection under varying budgets of $5$, $8$, and $10$ minutes and explore its effect on bug detection, confirming the advantage of smart selection over combining all goals.

cs.SE↗

arXiv2021

TFix+: Self-configuring Hybrid Timeout Bug Fixing for Cloud Systems

Timeout bugs can cause serious availability and performance issues which are often difficult to fix due to the lack of diagnostic information. Previous work proposed solutions for fixing specific type of timeout-related performance bugs. In this paper, we present TFix+, a self-configuring timeout bug fixing framework for automatically correcting two major kinds of timeout bugs (i.e., misused timeout bugs and missing timeout bugs) with dynamic timeout value predictions. TFix+ provides two new hybrid schemes for fixing misused and missing timeout bugs, respectively. TFix+ further provides prediction-driven timeout variable configuration based on runtime function tracing. We have implemented a prototype of TFix+ and conducted experiments on 16 real world timeout bugs. Our experimental results show that TFix+ can effectively fix 15 out of tested 16 timeout bugs.

cs.SE↗