arXiv subjects

Deqing Zou

Publications and source records attributed to Deqing Zou.

At least 19 recordsLinked to original sources

arXiv2026

Towards Security-Auditable LLM Agents: A Unified Graph Representation

LLM-based agentic systems are rapidly evolving to perform complex autonomous tasks through dynamic tool invocation, stateful memory management, and multi-agent collaboration. However, this semantics-driven execution paradigm creates a severe semantic gap between low-level physical events and high-level execution intent, making post-hoc security auditing fundamentally difficult. Existing representation mechanisms, including static SBOMs and runtime logs, provide only fragmented evidence and fail to capture cognitive-state evolution, capability bindings, persistent memory contamination, and cascading risk propagation across interacting agents. To bridge this gap, we propose Agent-BOM, a unified structural representation for agent security auditing. Agent-BOM models an agentic system as a hierarchical attributed directed graph that separates static capability bases, such as models, tools, and long-term memory, from dynamic runtime semantic states, such as goals, reasoning trajectories, and actions. These layers are connected through semantic edges and security attributes, transforming fragmented execution traces into queryable audit paths. Building on Agent-BOM, we develop a graph-query-based paradigm for path-level risk assessment and instantiate it with the OWASP Agentic Top 10. We further implement an auditing plugin in the OpenClaw environment to construct Agent-BOM from live executions. Evaluation on representative real-world agentic attack scenarios shows that Agent-BOM can reconstruct stealthy attack chains, including cross-session memory poisoning and tool misuse, capability supply-chain hijacking and unexpected code execution, multi-agent ecosystem hijacking, and privilege and trust abuse. These results demonstrate that Agent-BOM provides a unified and auditable foundation for root-cause analysis and security adjudication in complex agentic ecosystems.

cs.AI↗

arXiv2026

Mission (Im)Possible: Towards Autonomous Discovery of Autopilot Logic Flaws in Open-Source Drone Frameworks

Autonomous drones are increasingly deployed in safety-critical tasks, yet their autopilot software can still contain subtle logic flaws that lead to unsafe behaviors. These flaws are difficult to discover with existing techniques because they often depend on semantic relationships across modules, commands, and runtime state, and rarely manifest as crashes or detectable failures. With recent advances in LLMs, we observe that, when equipped with essential domain knowledge and carefully crafted context, LLMs demonstrate strong capabilities in understanding the high-level semantics of autopilot systems, making them suitable for logic flaw discovery. However, LLMs are limited by their context and can be biased to the vulnerabilities frequently occurs in training dataset, while logic flaws often require higher-level semantic bug oracles to identify their existence. To bridge the gap, we present Impossible Mission Finder (IMF), a multi-agent system combining static analysis, dynamic execution and log analysis, that supports fully automated identify, test, and validate logic flaws in autonomous vehicles. IMF primarily contains three LLM agents, it first uses the auditor agent to propose logic flaw hypothesis based on carefully-crafted program slicing, then it launches tester agents to write and execute test cases for each candidate, and finally it spawns the log analyzer agent to validate the execution log against the flaw hypothesis. This enables IMF dynamically validate bug oracles in a case-by-case manner, making it suitable for discovering logic flaws that do not have a unified pattern. We evaluated IMF on two widely used open-source autopilot systems, PX4 and Ardupilot, leading to 105 previously unknown logic flaws, that can cause unsafe behaviors including mission interruption failures, invalid actions, entering into restricted fly zones etc.

cs.CR↗

arXiv2026

Why Static Skill Inconsistencies Do Not Equate to Runtime Risks: A Behavioral Analysis of LLM Agent Execution

LLM agent frameworks heavily rely on external skill assets to standardize tool usage, encode operational workflows, and supply executable runtime resources. Despite their widespread adoption, multi-component skill packages frequently suffer from cross-artifact inconsistencies, where documentation, scripts, and configuration files specify conflicting behavioral constraints. Although static auditing techniques can uncover such contractual discrepancies, static defect volume cannot reliably reflect real-world agent performance degradation. This gap leaves practitioners without rigorous evidence to assess the practical risk of flawed agent skills. This paper presents a systematic empirical study to demystify the runtime implications of skill inconsistencies. We conduct large-scale auditing over 575 community and benchmark skills, yielding 1,182 manually validated contract inconsistencies. We further build a controlled evaluation pipeline with 49 executable agent tasks and three representative LLMs across NoSkill, OriginalSkill, and FixedSkill experimental settings. Through fine-grained behavioral analysis of 441 paired execution trajectories, we quantify the actual performance impacts of resolved defects.

cs.AI↗

arXiv2026

Bulkhead: Automated Semantic Detection and Remediation of Container Escape Vulnerabilities

Container filesystem isolation depends on safely resolving container-controlled paths during privileged host-container interactions. As third-party plugins and agent sandboxes introduce more interactions, unsafe path resolution can redirect host operations across isolation boundaries, expanding the attack surface for path traversal vulnerabilities. These logic flaws are difficult to mitigate: kernel defenses require intrusive redesign, static rules miss cross-boundary semantics, and manual auditing cannot scale. Large Language Model (LLM) based methods can detect syntax flaws through stable code patterns and identify logic flaws in domains such as smart contracts by comparing equivalent implementations. Container components, however, provide neither form of regularity: their heterogeneous interfaces operate under trust boundaries, filesystem views and interaction orders, making the security invariant difficult to find. We present Bulkhead, an automated framework for detection and remediation. Bulkhead summarizes historical CVEs into structured semantic patterns, correlates documentation with entry and effect side code evidence, and recovers source-to-sink execution contexts. Grounded patterns guide LLM reasoning, while isolated Proof-of-Concept (PoC) execution validates findings. For remediation, Bulkhead generates repository-specific patches, checks that they build, and iteratively verifies path safety and functionality using pattern derived model checking templates and counterexamples before runtime validation. Across container repositories, Bulkhead discovered 20 zero-day vulnerabilities, accounting for 43% of the total number discovered over the past decade. Among them, 3 have been assigned CVEs and the verified patch suggestions have been submitted to the upstream.

cs.CR↗

arXiv2026

Characterizing Security Effects of OSS Vulnerabilities in Agent Systems

Software agents increasingly depend on open-source components when executing tools and interacting with external systems. Security flaws in these dependencies may therefore influence more than the software process in which they occur: their consequences can be carried through tool outputs, agent state, and information subsequently exposed to the model. Determining whether such a consequence is actually realized in a particular execution, and where its influence stops within the agent system, remains challenging. We investigate how known OSS vulnerabilities behave when exercised as part of agent workflows. Our study reveals recurring patterns in the way security-relevant consequences emerge and propagate across runtime layers. Building on these observations, we use vulnerability-aware semantic information, differential executions of vulnerable and corrected software, and runtime provenance spanning multiple layers to determine whether a vulnerability produces an observable security effect and to identify the furthest layer at which that effect remains manifested. Our evaluation shows that this approach can accurately distinguish realized vulnerability effects and determine their manifestation boundaries across a diverse collection of vulnerability scenarios. We additionally apply the analysis to documented workflows in a real-world agent framework and uncover multiple security effects originating from known vulnerabilities in its OSS dependencies. These results highlight the importance of reasoning about vulnerable dependencies in terms of their execution-level consequences rather than vulnerability presence alone.

cs.SE↗

arXiv2026

AGATE: Provenance-Based Runtime Defense Against Compositional Attacks on LLM Agents

LLM agents can produce harmful effects through sequences of ordinary operations. Judging such actions requires establishing both the authority that permits them and the origin of the data they carry. We present AGATE, an authorization and data-provenance gate at instrumented agent-harness boundaries. Operator declarations and host approval events ground authorization; delegated actions are constrained by grants that bind to exact parameters, expire, and permit a limited number of uses. Source registration connects observed inputs to subsequent transfers, while an effect ledger tracks repeated requests. Deterministic checks make decisions without an LLM in the decision path and retain their grounds with execution evidence for forensic replay. Adapters integrate three production harnesses -- DeepSeek Harness, OpenCode, and OpenClaw -- without modifying host code, translating each host's native observation and veto points into a single shared gate interface; the judgment core is identical in all three, and only enforcement depth differs. Our evaluation combines 153 exercised attack-chain records with deployment, utility, and reconstruction experiments. The deployment observations expose how tool declarations and data checks govern business actions, including a bypass through parameter rewriting. Six of eleven benign file-processing scenarios contain denial events, revealing the utility cost of content-based provenance policies. Across 252 runs on 63 sanitized scenarios, replay agrees with live graph projections for all 63 scenarios on each of two platforms. These results establish the feasibility of provenance-based runtime judgment and identify content transformation, legitimate reuse, and observation coverage as concrete limits.

cs.CR↗

arXiv2026

From Component Snapshots to Lifecycle Traces: Agent-Based Software Composition Analysis

Software supply-chain security requires accurate identification of third-party components and an understanding of how they evolve from development to execution. Existing software composition analysis (SCA) approaches examine manifests, build environments, release artifacts, containers, or runtime states, but typically produce only stage-specific views of software composition. As dependencies are resolved, removed, repackaged, and transformed across lifecycle stages, a single snapshot cannot capture both where a component originates and where it ultimately ends up. Combining snapshots from multiple stages still leaves their cross-stage relationships unresolved. We present SCA-Agent, an agent-based approach to lifecycle-aware SCA that reconstructs evidence-backed component lifecycle traces across Code, Build, Release, Deploy, and Runtime. SCA-Agent adaptively explores project-specific analysis paths, gathers stage-specific evidence, and correlates observations across stages to recover component identities, versions, introduction paths, propagation relationships, and final lifecycle states. We evaluate SCA-Agent on 105 real-world projects from the Java, JavaScript, and Python ecosystems. SCA-Agent achieves the highest component detection F1 across all lifecycle stages and ecosystems. For vulnerability exposure assessment, it reaches an F1 score of 96.69%, exceeding the best traditional SCA tool by 18.76 percentage points. These results show that lifecycle-aware SCA supports traceable component provenance and more accurate software supply-chain risk assessment.

cs.SE↗

arXiv2026

A Survey of Large Audio Language Models: Generalization, Trustworthiness, and Outlook

Advances in Large Language Models (LLMs) have paved the way for Multimodal Large Language Models (MLLMs). Among these, Large Audio Language Models (LALMs) are essential for realizing universal auditory intelligence. Despite their remarkable performance, the escalation of LALMs' capabilities has significantly outpaced the development of systemic frameworks to ensure their trustworthiness. This survey provides a comprehensive investigation into the endogenous mechanisms of LALMs, detailing the architectural innovations and alignment algorithms that facilitate emergent reasoning. Specifically, we analyze how the transition to unified end-to-end frameworks and the integration of continuous acoustic signals expand the attack surface. To rigorously evaluate the risks within these paradigms, we establish a comprehensive taxonomy of trustworthiness, categorizing critical vulnerabilities such as cross-modal jailbreaking, latent acoustic backdoors, and biometric privacy leakage. We review the state-of-the-art LALMs through six analytical pillars: hallucination, robustness, safety, privacy, fairness, and authentication. The pronounced imbalance between a mature offensive landscape and underdeveloped defenses highlights persistent trustworthiness gaps and multidimensional risks in audio-centric intelligence. Finally, we propose a roadmap advocating for ``Defense-in-Depth'' architectures, causal auditory world modeling, and intrinsic representation engineering to support the development of more reliable and trustworthy audio intelligence. Our project has been uploaded to GitHub https://github.com/Kwwwww74/Awesome-Trustworthy-AudioLLMs.

cs.SD↗

arXiv2026

MemArbiter: Decision-Time Memory Arbitration for Long-Horizon LLM Agents

Large language model (LLM) agents must retain and use cross-step information to act coherently in long-horizon tasks. Existing methods improve memory accessibility, yet action-relevant information may still fail to guide the current decision because it is poorly formed, organized, prioritized, or presented. We call this post-access failure the Memory-Action Gap. We propose MemArbiter, a function-aware memory arbitration framework that addresses the memory-management-induced component of this gap. MemArbiter decomposes interaction histories into atomic items, organizes them into five functional Memory Banks, and combines bank-level demand, item-level relevance, focal-ambient representations, and a temporal presentation gate to dynamically control memory salience. We evaluate MemArbiter on ALFWorld against Flat Retrieval and Flat Recency under unified per-step memory budgets. With an open-weight action-generation model, MemArbiter achieves success rates of 82.8% and 92.5% under 500- and 750-token budgets, outperforming the strongest baseline by 20.9 and 25.4 percentage points, respectively. It also improves post-failure recovery and reduces failed-action repetition and state-action recurrence. These results show that function-aware memory arbitration enables accessible information to guide actions more effectively.

cs.AI↗

arXiv2025

Semantically-Equivalent Transformations-Based Backdoor Attacks against Neural Code Models: Characterization and Mitigation

Neural code models have been increasingly incorporated into software development processes. However, their susceptibility to backdoor attacks presents a significant security risk. The state-of-the-art understanding focuses on injection-based attacks, which insert anomalous patterns into software code. These attacks can be neutralized by standard sanitization techniques. This status quo may lead to a false sense of security regarding backdoor attacks. In this paper, we introduce a new kind of backdoor attacks, dubbed Semantically-Equivalent Transformation (SET)-based backdoor attacks, which use semantics-preserving low-prevalence code transformations to generate stealthy triggers. We propose a framework to guide the generation of such triggers. Our experiments across five tasks, six languages, and models like CodeBERT, CodeT5, and StarCoder show that SET-based attacks achieve high success rates (often >90%) while preserving model utility. The attack proves highly stealthy, evading state-of-the-art defenses with detection rates on average over 25.13% lower than injection-based counterparts. We evaluate normalization-based countermeasures and find they offer only partial mitigation, confirming the attack's robustness. These results motivate further investigation into scalable defenses tailored to SET-based attacks.

cs.SE↗

arXiv2025

FEANEL: A Benchmark for Fine-Grained Error Analysis in K-12 English Writing

Large Language Models (LLMs) have transformed artificial intelligence, offering profound opportunities for educational applications. However, their ability to provide fine-grained educational feedback for K-12 English writing remains underexplored. In this paper, we challenge the error analysis and pedagogical skills of LLMs by introducing the problem of Fine-grained Error Analysis for English Learners and present the Fine-grained Error ANalysis for English Learners (FEANEL) Benchmark. The benchmark comprises 1,000 essays written by elementary and secondary school students, and a well-developed English writing error taxonomy. Each error is annotated by language education experts and categorized by type, severity, and explanatory feedback, using a part-of-speech-based taxonomy they co-developed. We evaluate state-of-the-art LLMs on the FEANEL Benchmark to explore their error analysis and pedagogical abilities. Experimental results reveal significant gaps in current LLMs' ability to perform fine-grained error analysis, highlighting the need for advancements in particular methods for educational applications.

cs.CL↗

arXiv2025

Position: LLMs Can be Good Tutors in English Education

While recent efforts have begun integrating large language models (LLMs) into English education, they often rely on traditional approaches to learning tasks without fully embracing educational methodologies, thus lacking adaptability to language learning. To address this gap, we argue that LLMs have the potential to serve as effective tutors in English Education. Specifically, LLMs can play three critical roles: (1) as data enhancers, improving the creation of learning materials or serving as student simulations; (2) as task predictors, serving as learner assessment or optimizing learning pathway; and (3) as agents, enabling personalized and inclusive education. We encourage interdisciplinary research to explore these roles, fostering innovation while addressing challenges and risks, ultimately advancing English Education through the thoughtful integration of LLMs.

cs.CL↗

arXiv2025

SoK: Automated Vulnerability Repair: Methods, Tools, and Assessments

The increasing complexity of software has led to the steady growth of vulnerabilities. Vulnerability repair investigates how to fix software vulnerabilities. Manual vulnerability repair is labor-intensive and time-consuming because it relies on human experts, highlighting the importance of Automated Vulnerability Repair (AVR). In this SoK, we present the systematization of AVR methods through the three steps of AVR workflow: vulnerability analysis, patch generation, and patch validation. We assess AVR tools for C/C++ and Java programs as they have been widely studied by the community. Since existing AVR tools for C/C++ programs are evaluated with different datasets, which often consist of a few vulnerabilities, we construct the first C/C++ vulnerability repair benchmark dataset, dubbed Vul4C, which contains 144 vulnerabilities as well as their exploits and patches. We use Vul4C to evaluate seven AVR tools for C/C++ programs and use the third-party Vul4J dataset to evaluate two AVR tools for Java programs. We also discuss future research directions.

cs.SE↗

arXiv2025

Revisiting Classification Taxonomy for Grammatical Errors

Grammatical error classification plays a crucial role in language learning systems, but existing classification taxonomies often lack rigorous validation, leading to inconsistencies and unreliable feedback. In this paper, we revisit previous classification taxonomies for grammatical errors by introducing a systematic and qualitative evaluation framework. Our approach examines four aspects of a taxonomy, i.e., exclusivity, coverage, balance, and usability. Then, we construct a high-quality grammatical error classification dataset annotated with multiple classification taxonomies and evaluate them grounding on our proposed evaluation framework. Our experiments reveal the drawbacks of existing taxonomies. Our contributions aim to improve the precision and effectiveness of error analysis, providing more understandable and actionable feedback for language learners.

cs.CL↗

arXiv2024

On the Effectiveness of Function-Level Vulnerability Detectors for Inter-Procedural Vulnerabilities

Software vulnerabilities are a major cyber threat and it is important to detect them. One important approach to detecting vulnerabilities is to use deep learning while treating a program function as a whole, known as function-level vulnerability detectors. However, the limitation of this approach is not understood. In this paper, we investigate its limitation in detecting one class of vulnerabilities known as inter-procedural vulnerabilities, where the to-be-patched statements and the vulnerability-triggering statements belong to different functions. For this purpose, we create the first Inter-Procedural Vulnerability Dataset (InterPVD) based on C/C++ open-source software, and we propose a tool dubbed VulTrigger for identifying vulnerability-triggering statements across functions. Experimental results show that VulTrigger can effectively identify vulnerability-triggering statements and inter-procedural vulnerabilities. Our findings include: (i) inter-procedural vulnerabilities are prevalent with an average of 2.8 inter-procedural layers; and (ii) function-level vulnerability detectors are much less effective in detecting to-be-patched functions of inter-procedural vulnerabilities than detecting their counterparts of intra-procedural vulnerabilities.

cs.CR↗

arXiv2023

Robin: A Novel Method to Produce Robust Interpreters for Deep Learning-Based Code Classifiers

Deep learning has been widely used in source code classification tasks, such as code classification according to their functionalities, code authorship attribution, and vulnerability detection. Unfortunately, the black-box nature of deep learning makes it hard to interpret and understand why a classifier (i.e., classification model) makes a particular prediction on a given example. This lack of interpretability (or explainability) might have hindered their adoption by practitioners because it is not clear when they should or should not trust a classifier's prediction. The lack of interpretability has motivated a number of studies in recent years. However, existing methods are neither robust nor able to cope with out-of-distribution examples. In this paper, we propose a novel method to produce \underline{Rob}ust \underline{in}terpreters for a given deep learning-based code classifier; the method is dubbed Robin. The key idea behind Robin is a novel hybrid structure combining an interpreter and two approximators, while leveraging the ideas of adversarial training and data augmentation. Experimental results show that on average the interpreter produced by Robin achieves a 6.11\% higher fidelity (evaluated on the classifier), 67.22\% higher fidelity (evaluated on the approximator), and 15.87x higher robustness than that of the three existing interpreters we evaluated. Moreover, the interpreter is 47.31\% less affected by out-of-distribution examples than that of LEMNA.

cs.SE↗

arXiv2022

Contrastive Learning for Robust Android Malware Familial Classification

Due to its open-source nature, Android operating system has been the main target of attackers to exploit. Malware creators always perform different code obfuscations on their apps to hide malicious activities. Features extracted from these obfuscated samples through program analysis contain many useless and disguised features, which leads to many false negatives. To address the issue, in this paper, we demonstrate that obfuscation-resilient malware family analysis can be achieved through contrastive learning. The key insight behind our analysis is that contrastive learning can be used to reduce the difference introduced by obfuscation while amplifying the difference between malware and other types of malware. Based on the proposed analysis, we design a system that can achieve robust and interpretable classification of Android malware. To achieve robust classification, we perform contrastive learning on malware samples to learn an encoder that can automatically extract robust features from malware samples. To achieve interpretable classification, we transform the function call graph of a sample into an image by centrality analysis. Then the corresponding heatmaps can be obtained by visualization techniques. These heatmaps can help users understand why the malware is classified as this family. We implement \emph{IFDroid} and perform extensive evaluations on two datasets. Experimental results show that \emph{IFDroid} is superior to state-of-the-art Android malware familial classification systems. Moreover, \emph{IFDroid} is capable of maintaining a 98.4\% F1 on classifying 69,421 obfuscated malware samples.

cs.CR↗

arXiv2021

Towards Making Deep Learning-based Vulnerability Detectors Robust

Automatically detecting software vulnerabilities in source code is an important problem that has attracted much attention. In particular, deep learning-based vulnerability detectors, or DL-based detectors, are attractive because they do not need human experts to define features or patterns of vulnerabilities. However, such detectors' robustness is unclear. In this paper, we initiate the study in this aspect by demonstrating that DL-based detectors are not robust against simple code transformations, dubbed attacks in this paper, as these transformations may be leveraged for malicious purposes. As a first step towards making DL-based detectors robust against such attacks, we propose an innovative framework, dubbed ZigZag, which is centered at (i) decoupling feature learning and classifier learning and (ii) using a ZigZag-style strategy to iteratively refine them until they converge to robust features and robust classifiers. Experimental results show that the ZigZag framework can substantially improve the robustness of DL-based detectors.

cs.CR↗