arXiv subjects

Yifei Ge

Publications and source records attributed to Yifei Ge.

11 recordsLinked to original sources

arXiv2026

Execution-Grounded Security Testing for Coding Agents in Software Engineering Pipelines

Coding agents are increasingly integrated into system operations, where their tool use can directly modify project artifacts, execution environments, and the underlying system. For example, if a coding agent inserts a hook into a system startup or configuration script, that change can persist after the interaction, be triggered later, and abuse delegated user or system privileges to modify the system. This makes security testing a system problem: the key question is not only what the agent says, but what it actually does to the surrounding environment. We present an execution-grounded red-team testing framework for probing this execution-layer security boundary using observable sandbox evidence, including tool invocations, runtime traces, and file-system diffs. Our framework embeds target unsafe operations into routine software engineering workloads, including unit testing, regression testing, crash reproduction, and validation, and uses an execution oracle to guide refinement when an initial probe is rejected or fails. Across multiple agent frameworks and model backbones, our red-team workload reformulation substantially increases verified unsafe execution, reaching 73.61% on code carriers and 53.93% on text carriers. These results show that coding agents in system operations remain insecure under task disguise: once risky intent is hidden inside plausible engineering tasks, the agent can be induced to carry out unsafe actions on the surrounding system. More broadly, coding agents in system operations still demand stronger security testing and safeguards.

cs.AI↗

arXiv2026

Probing Privacy Leaks in LLM-based Code Generation via Test Generation

The widespread availability of large-scale code datasets has fueled the rapid development of large language models (LLMs) for code-related tasks. These datasets may include sensitive personally identifiable information (PII), which can lead to privacy leakage when LLMs memorize and reproduce it. However, existing privacy-leakage detection methods rely on ad-hoc prompt construction (manually or automatically designed). Therefore, they do not adequately approximate the real-world contexts in which PII appears in code corpora, making it difficult to extract realistic privacy leakage. In this paper, we propose a pipeline that simulates practical privacy-related code generation scenarios and adopts a test-driven strategy to elicit the memorized information from the generated test cases. We further introduce an automatically constructed privacy feature library that replaces manual prompt engineering by providing realistic templates and examples to guide test case generation. Large-scale experiments on 5 widely used LLMs show that our pipeline exposes more confirmed privacy leakage, achieving a 2.56 times increase in detected leakage compared to existing baselines.

cs.SE↗

arXiv2024

Security of Language Models for Code: A Systematic Literature Review

Language models for code (CodeLMs) have emerged as powerful tools for code-related tasks, outperforming traditional methods and standard machine learning approaches. However, these models are susceptible to security vulnerabilities, drawing increasing research attention from domains such as software engineering, artificial intelligence, and cybersecurity. Despite the growing body of research focused on the security of CodeLMs, a comprehensive survey in this area remains absent. To address this gap, we systematically review 67 relevant papers, organizing them based on attack and defense strategies. Furthermore, we provide an overview of commonly used language models, datasets, and evaluation metrics, and highlight open-source tools and promising directions for future research in securing CodeLMs.

cs.SE↗

arXiv2024

Demonstration Attack against In-Context Learning for Code Intelligence

Recent advancements in large language models (LLMs) have revolutionized code intelligence by improving programming productivity and alleviating challenges faced by software developers. To further improve the performance of LLMs on specific code intelligence tasks and reduce training costs, researchers reveal a new capability of LLMs: in-context learning (ICL). ICL allows LLMs to learn from a few demonstrations within a specific context, achieving impressive results without parameter updating. However, the rise of ICL introduces new security vulnerabilities in the code intelligence field. In this paper, we explore a novel security scenario based on the ICL paradigm, where attackers act as third-party ICL agencies and provide users with bad ICL content to mislead LLMs outputs in code intelligence tasks. Our study demonstrates the feasibility and risks of such a scenario, revealing how attackers can leverage malicious demonstrations to construct bad ICL content and induce LLMs to produce incorrect outputs, posing significant threats to system security. We propose a novel method to construct bad ICL content called DICE, which is composed of two stages: Demonstration Selection and Bad ICL Construction, constructing targeted bad ICL content based on the user query and transferable across different query inputs. Ultimately, our findings emphasize the critical importance of securing ICL mechanisms to protect code intelligence systems from adversarial manipulation.

cs.CR↗

arXiv2024

The Milky Way Atlas for Linear Filaments

Filamentary structure is important for the ISM and star formation. Galactic distribution of filaments may regulate the star formation rate in the Milky Way. However, interstellar filaments are intrinsically complex, making it difficult to study quantitatively. Here, we focus on linear filaments, the simplest morphology that can be treated as building blocks of any filamentary structure. We present the first catalog of 42 ``straight-line'' filaments across the full Galactic plane, identified by clustering of far-IR Herschel HiGAL clumps in position-position-velocity space. We use molecular line cubes to investigate the dynamics along the filaments; compare the filaments with Galactic spiral arms; and compare ambient magnetic fields with the filaments' orientation. The selected filaments show extreme linearity ($>$10), aspect ratio (7-48), and velocity coherence over a length of 3-40 pc (mostly $>$10 pc). About 1/3 of them are associated with spiral arms, but only one is located in arm center, a.k.a. ``bones'' of the Milky Way. A few of them extend perpendicular to the Galactic plane, and none is located in the Central Molecular Zone (CMZ) near the Galactic center. Along the filaments, prevalent periodic oscillation (both in velocity and density) is consistent with gas flows channeled by the filaments and feeding the clumps which harbor diverse star formation activities. No correlation is found between the filament orientations with Planck measured global magnetic field lines. This work highlights some of the fundamental properties of molecular filaments and provides a golden sample for follow-up studies on star formation, ISM structure, and Milky Way structure.

astro-ph.GA↗

arXiv2024

Renyi Differential Privacy in the Shuffle Model: Enhanced Amplification Bounds

The shuffle model of Differential Privacy (DP) has gained significant attention in privacy-preserving data analysis due to its remarkable tradeoff between privacy and utility. It is characterized by adding a shuffling procedure after each user's locally differentially private perturbation, which leads to a privacy amplification effect, meaning that the privacy guarantee of a small level of noise, say $\epsilon_0$, can be enhanced to $O(\epsilon_0/\sqrt{n})$ (the smaller, the more private) after shuffling all $n$ users' perturbed data. Most studies in the shuffle DP focus on proving a tighter privacy guarantee of privacy amplification. However, the current results assume that the local privacy budget $\epsilon_0$ is within a limited range. In addition, there remains a gap between the tightest lower bound and the known upper bound of the privacy amplification. In this work, we push forward the state-of-the-art by making the following contributions. Firstly, we present the first asymptotically optimal analysis of Renyi Differential Privacy (RDP) in the shuffle model without constraints on $\epsilon_0$. Secondly, we introduce hypothesis testing for privacy amplification through shuffling, offering a distinct analysis technique and a tighter upper bound. Furthermore, we propose a DP-SGD algorithm based on RDP. Experiments demonstrate that our approach outperforms existing methods significantly at the same privacy level.

math.CO↗

arXiv2023

A Generalized Shuffle Framework for Privacy Amplification: Strengthening Privacy Guarantees and Enhancing Utility

The shuffle model of local differential privacy is an advanced method of privacy amplification designed to enhance privacy protection with high utility. It achieves this by randomly shuffling sensitive data, making linking individual data points to specific individuals more challenging. However, most existing studies have focused on the shuffle model based on $(\epsilon_0,0)$-Locally Differentially Private (LDP) randomizers, with limited consideration for complex scenarios such as $(\epsilon_0,\delta_0)$-LDP or personalized LDP (PLDP). This hinders a comprehensive understanding of the shuffle model's potential and limits its application in various settings. To bridge this research gap, we propose a generalized shuffle framework that can be applied to any $(\epsilon_i,\delta_i)$-PLDP setting with personalized privacy parameters. This generalization allows for a broader exploration of the privacy-utility trade-off and facilitates the design of privacy-preserving analyses in diverse contexts. We prove that shuffled $(\epsilon_i,\delta_i)$-PLDP process approximately preserves $\mu$-Gaussian Differential Privacy with \mu = \sqrt{\frac{2}{\sum_{i=1}^{n} \frac{1-\delta_i}{1+e^{\epsilon_i}}-\max_{i}{\frac{1-\delta_{i}}{1+e^{\epsilon_{i}}}}}}. $ This approach allows us to avoid the limitations and potential inaccuracies associated with inequality estimations. To strengthen the privacy guarantee, we improve the lower bound by utilizing hypothesis testing} instead of relying on rough estimations like the Chernoff bound or Hoeffding's inequality. Furthermore, extensive comparative evaluations clearly show that our approach outperforms existing methods in achieving strong central privacy guarantees while preserving the utility of the global model. We have also carefully designed corresponding algorithms for average function, frequency estimation, and stochastic gradient descent.

cs.CR↗

arXiv2023

A Survey of Source Code Search: A 3-Dimensional Perspective

(Source) code search is widely concerned by software engineering researchers because it can improve the productivity and quality of software development. Given a functionality requirement usually described in a natural language sentence, a code search system can retrieve code snippets that satisfy the requirement from a large-scale code corpus, e.g., GitHub. To realize effective and efficient code search, many techniques have been proposed successively. These techniques improve code search performance mainly by optimizing three core components, including query understanding component, code understanding component, and query-code matching component. In this paper, we provide a 3-dimensional perspective survey for code search. Specifically, we categorize existing code search studies into query-end optimization techniques, code-end optimization techniques, and match-end optimization techniques according to the specific components they optimize. Considering that each end can be optimized independently and contributes to the code search performance, we treat each end as a dimension. Therefore, this survey is 3-dimensional in nature, and it provides a comprehensive summary of each dimension in detail. To understand the research trends of the three dimensions in existing code search studies, we systematically review 68 relevant literatures. Different from existing code search surveys that only focus on the query end or code end or introduce various aspects shallowly (including codebase, evaluation metrics, modeling technique, etc.), our survey provides a more nuanced analysis and review of the evolution and development of the underlying techniques used in the three ends. Based on a systematic review and summary of existing work, we outline several open challenges and opportunities at the three ends that remain to be addressed in future work.

cs.SE↗

arXiv2023

An Extractive-and-Abstractive Framework for Source Code Summarization

(Source) Code summarization aims to automatically generate summaries/comments for a given code snippet in the form of natural language. Such summaries play a key role in helping developers understand and maintain source code. Existing code summarization techniques can be categorized into extractive methods and abstractive methods. The extractive methods extract a subset of important statements and keywords from the code snippet using retrieval techniques, and generate a summary that preserves factual details in important statements and keywords. However, such a subset may miss identifier or entity naming, and consequently, the naturalness of generated summary is usually poor. The abstractive methods can generate human-written-like summaries leveraging encoder-decoder models from the neural machine translation domain. The generated summaries however often miss important factual details. To generate human-written-like summaries with preserved factual details, we propose a novel extractive-and-abstractive framework. The extractive module in the framework performs a task of extractive code summarization, which takes in the code snippet and predicts important statements containing key factual details. The abstractive module in the framework performs a task of abstractive code summarization, which takes in the entire code snippet and important statements in parallel and generates a succinct and human-written-like natural language summary. We evaluate the effectiveness of our technique, called EACS, by conducting extensive experiments on three datasets involving six programming languages. Experimental results show that EACS significantly outperforms state-of-the-art techniques in terms of all three widely used metrics, including BLEU, METEOR, and ROUGH-L.

cs.SE↗

arXiv2023

The ALMA Survey of Star Formation and Evolution in Massive Protoclusters with Blue Profiles (ASSEMBLE): Core Growth, Cluster Contraction, and Primordial Mass Segregation

The ALMA Survey of Star Formation and Evolution in Massive Protoclusters with Blue Profiles (ASSEMBLE) aims to investigate the process of mass assembly and its connection to high-mass star formation theories in protoclusters in a dynamic view. We observed 11 massive (Mclump>1000 Msun), luminous (Lbol>10,000 Lsun), and blue-profile (infall signature) clumps by ALMA with resolution of 2200-5500 au at 350 GHz (870 um) in continuum and line emission. 248 dense cores were identified, including 106 cores showing protostellar signatures and 142 prestellar core candidates. Compared to early-stage infrared dark clouds (IRDCs) by ASHES, the core mass and surface density within the ASSEMBLE clumps exhibited significant increment, suggesting concurrent core accretion during the evolution of the clumps. The maximum mass of prestellar cores was found to be 2 times larger than that in IRDCs, indicating evolved protoclusters have the potential to harbor massive prestellar cores. The mass relation between clumps and their most massive core (MMCs) is observed in ASSEMBLE but not in IRDCs, which is suggested to be regulated by multiscale mass accretion. The mass correlation between the core clusters and their MMCs has a steeper slope compared to that observed in stellar clusters, which can be due to fragmentation of the MMC and stellar multiplicity. We observe a decrease in core separation and an increase in central concentration as protoclusters evolve. We confirm primordial mass segregation in the ASSEMBLE protoclusters, possibly resulting from gravitational concentration and/or gas accretion.

astro-ph.GA↗

arXiv2022

A census of 163 large-scale (>=10 pc), velocity-coherent filaments in inner Galactic plane: physical properties, dense gas fraction, and association with spiral arms

The interstellar medium has a highly filamentary and hierarchical structure, which may play a significant role in star formation. A systematical study on the large-scale filaments towards their physical parameters, distribution, structures and kinematics will inform us of what kind of filaments have potential to form stars, how the material feed protostars through filaments, and the connection between star formation and Galactic spiral arms. Unlike the traditional "by eyes" searches, we use a customized minimum spanning tree algorithm to identify filaments by linking Galactic clumps from the APEX Telescope Large Area Survey of the Galaxy catalogue. In the inner Galactic plane ($|l| < 60^\circ$), we identify 163 large-scale filaments with physical properties derived, including dense gas mass fraction, and compare them with an updated spiral arm model in position-position-velocity space. Dense gas mass fraction is found not to differ significantly in various Galactic position, neither does it in different spiral arms. We also find that most filaments are inter-arm filaments after adding a distance constraint, and filaments in arm differ a little with those not in. One surprising result is that clumps on and off filaments have no significant distinction in their mass at the same size.

astro-ph.GA↗