Searcharxiv⌕ Search

arXiv subjects

Zhiqiang Lin

Publications and source records attributed to Zhiqiang Lin.

At least 19 recordsLinked to original sources
arXiv2026

Association-based Privacy Attacks in Wireless Protocols: Formal Modeling and Mitigation

With the surge in privacy-sensitive data from sources such as social media and IoT devices, there is a pressing need for formal, automated methods to assess privacy risks within these intricate systems. This paper formally investigates root sources of pairing-based privacy threats exploited using replay/relay techniques in wireless communication. Our research harnesses condition-oblivious responses, replay-resistance, and distance bounding measures vital for protocols utilizing shared keys in allowlists for authenticated reconnections. Particularly, the paper uses formal modeling of notable wireless networks, like the Wi-Fi P2P persistent group formation and the Bluetooth Low Energy reconnection procedure, to illustrate the root causes and countermeasures. Our model rigorously validates the proposed solution against association inference attacks, along with existing formalizations of well-authentication, frame opacity, and no-desynchronization. The ensuing analysis reveals not only uncharted privacy realms in wireless communication but also identifies old and new vulnerabilities. Our proposed design changes are acknowledged by Wi-Fi Alliance and Bluetooth SIG, paving the way for future advancements in resilient, privacy-preserving wireless protocols.

cs.CR↗
arXiv2026

Abstractions for Network Intelligence: A Reference Architecture for AI at the Wireless Edge

Networks are increasingly adopting AI as are AI applications leveraging networks. Awareness sharing between networks and AI applications promises to unlock higher levels of network utilization and application performance, but is inadequately supported in the current architecture of the Internet. In this paper, we describe a reference architecture that abstractly enables the synergistic interaction of intelligent applications and the intelligent network, via an information waist, and also supports the network intelligence services in the emerging intelligence plane in networks. We discuss the rationale for our AI-EDGE architecture, its functional requirements, and the core abstractions. We present a reference component-level design of the core abstractions to support experimentation and development on existing platforms for wireless networking (i.e., based on O-RAN cellular networks) and edge computing (i.e., based on 3GPP Edge App and ETSI MEC). Moreover, we provide representative use cases from the perspective of different types of users that demonstrate the benefits of the architecture in contexts of awareness sharing, portability, prototyping, and validation.

cs.NI↗
arXiv2026

CyberChainBench: Can AI Agents Secure Smart Contracts Against Real-World On-Chain Vulnerabilities?

We present CyberChainBench, a benchmark for evaluating LLM-based agents on smart contract security across three complementary tasks: vulnerability detection, exploit generation, and patch synthesis. Built from 541 real-world exploit incidents from DeFiHackLabs spanning 9 EVM chains, the benchmark provides end-to-end on-chain evaluation where agents interact with historical blockchain state through isolated evaluation environments orchestrated by Harbor, using tools to read code, trace transactions, and validate exploits on mainnet forks. Each case is anchored to a specific block and includes structured ground truth covering vulnerability type, localization, and attacker profit. Exploits are graded by economic impact on historical forks; patches are validated by replaying historical attacks and legitimate transactions as fail-to-pass test oracles on a proxy-upgradeable subset. We define a five-type vulnerability taxonomy and evaluate multiple agent--model configurations. Results reveal a clear difficulty gradient: the best configuration scores 37.5% on detection, 43.7% on exploitation, but only 23.4% on patching, with the top agent (Codex with GPT-5.5) realizing \$57.4M in total exploit profit across the 200-case exploit set at a cost of $2.39 per case.

cs.CR↗
arXiv2026

One Goal, Many Commands: Characterizing Denylist Fragility in AI Agents

The adoption of AI agents is increasing rapidly. Terminal AI agents, i.e., AI agents that run in terminal environments, are a widely used type of AI agents. Terminal AI agents rely heavily on shell command execution to interact with the host systems. They adopt a three-list command-gating mechanism to mitigate security risks introduced by command execution, with denylists serving as the load-bearing component. However, modern operating systems often ship a large, ever-expanding set of shell commands with complex functionalities. Our observation is that even a built-in denylist of Claude Code, well-maintained by its developers, can overlook bypass commands that invalidate its effectiveness. Such negligence leads to fragile command denylists that cannot even block operations that practitioners expect them to block. This paper presents the first systematic characterization of command denylist fragility in terminal AI agents. The paper formalizes the command denylist fragility problem and proposes an LLM-driven pipeline, ShellSieve, to detect such fragility. It prompts the LLM to propose possible bypasses and iteratively repairs them using feedback from a validator that executes them in a sandbox. In the evaluation, we applied ShellSieve to 1,709 real-world command denylists (containing 13,332 denylist rules) collected from GitHub. The evaluation shows several key findings, including that 69.0--98.6% of the denylists are fragile, that this fragility occurs consistently across projects and agents, and the validity of several possible root causes for this fragility. Our pipeline and findings will hopefully facilitate future research and practice regarding the command denylists used by AI agents.

cs.CR↗
arXiv2026

OpenPCC: Open and Confidential LLM Serving on Commodity TEEs

Generative AI applications such as personal AI agents, image generators, and chat assistants offer advanced capabilities to improve user experience. Behind the scenes, Large Language Models (LLMs) that power these services require a massive amount of computation and are usually deployed in the cloud, available as APIs, meaning that a user's request has to be sent to a Cloud Inference Service (CIS) for processing. However, the strong capabilities of LLM also mean that user's requests now contain much more personal sensitive or enterprise confidential information, demanding equally strong protection in CIS. While early industry efforts such as Apple Private Cloud Compute (PCC) and Google Private AI Compute have emerged to show the potential of secure CIS, they are not adoptable for deployment by others due to their reliance on proprietary hardware and closed ecosystem. In addition, they all suffer from their own design glitches that can undermine the ambitious goal of bringing in true privacy protection to end users. In this paper, we present our analysis of the fundamental requirements of building a secure yet open CIS. We then present OpenPCC, a Confidential CIS framework that does not rely on proprietary hardware but instead uses commercially available TEEs. We implement an open-source prototype and characterize it end-to-end on a Llama-3 8B vLLM workload, separating OpenPCC's own cost from the underlying TEE hardware. Our analysis and evaluation demonstrated the feasibility and security of the system.

cs.CR↗
arXiv2026

Confused ChatGPT: Cross-App Context Poisoning via First-Party APIs

ChatGPT Apps, launched by OpenAI on Oct. 6, 2025, introduce an app-in-app paradigm in which third-party applications share a single chat context with the user and with every other connected app. The ecosystem grew from 122 apps in Dec. 2025 to 888 by May 2026, yet its security has remained uninvestigated. We identify cross-app context poisoning, a variant of indirect prompt injection distinguished by three properties: 1) the injection persists in the shared chat context across turns; 2) the effect surfaces through a different co-resident app the user later invokes; and 3) the delivery vectors are first-party APIs exposed to every connected app. We find multiple APIs capable of writing app-controlled content into the shared context, with sendFollowUpMessage as the most direct and potent channel. Two undocumented parameters that the runtime silently accepts, systemPrompt and isVisible, amplify this channel to silent, system-priority writes. Leveraging this channel, we realize a confused-deputy attack in which a malicious app poisons the context so that the LLM, consulting that context, enables manipulation against benign co-resident apps. We demonstrate two payload styles (conditional and imperative) and evaluate them across six current ChatGPT models. The root cause is architectural: the LLM's context is a persistent, flat, untagged data store shared by user and apps, with no isolation. Every mature multi-tenant platform, from Multics virtual memory to Android UIDs and iOS sandbox profiles, paid the isolation cost before admitting third parties; ChatGPT Apps did not. Fixing this requires an architectural change, not a patch. We disclosed our findings to OpenAI; the undocumented parameters remain accessible at the time of writing, and the architectural gap is by design: the shared context that enables cross-app composition is the same flat namespace that enables cross-app poisoning.

cs.CR↗
arXiv2026

Source or It Didn't Happen: A Multi-Agent Framework for Citation Hallucination Detection

Large language models are increasingly used in scientific writing, yet they can fabricate citation-shaped references that appear plausible but fail bibliographic verification. Existing detectors often reduce verification to binary found/not-found decisions and rely on brittle parsing or incomplete retrieval, offering little field-level signal to auditors. We reframe citation hallucination detection as taxonomy-aligned field-level adjudication and introduce a 12-code taxonomy spanning Real, Potential, and Hallucinated citations. Based on this taxonomy, we build CiteTracer, a cascading multi-agent detector that extracts structured citations from PDF and BibTeX, retrieves evidence through cache lookup, URL fetch, scholar connectors, and web search, applies deterministic field matching, and routes ambiguous cases to class-specialist judgers. We release a benchmark of 2,450 synthetic citations built from real seeds with controlled LLM mutations, paired with 957 real-world fabricated citations drawn from ICLR 2026 and an anonymous conference desk-rejected submissions. CiteTracer reaches 97.1% accuracy on the synthetic benchmark, with class-level F1 scores of 97.0, 95.8, and 98.5 for Real, Potential, and Hallucinated, respectively, and detects 97.1% of fabrications on the real-world set without abstaining. Code: https://github.com/aaFrostnova/CiteTracer.

cs.CL↗
arXiv2026

REBENCH: A Procedural, Fair-by-Construction Benchmark for LLMs on Stripped-Binary Types and Names (Extended Version)

Large Language Models (LLMs) have achieved remarkable progress in recent years, driving their adoption across a wide range of domains, including computer security. In reverse engineering, LLMs are increasingly applied to critical tasks such as function and variable name recovery and type inference. However, despite the rapid growth of research in this area, progress has been hindered by the absence of a standardized dataset. Existing studies rely on disparate datasets, preprocessing pipelines, and evaluation metrics, making fair comparisons between approaches difficult and obscuring a clear understanding of LLM capabilities in binary analysis. To address these challenges, we present REBench, a comprehensive benchmark dataset for evaluating LLMs on binary reverse engineering tasks. REBench consolidates a superset of existing datasets, comprising hundreds of millions of lines of source code and a diverse collection of binaries spanning multiple architectures and optimization levels. REBench adopts a knowledge-base-driven methodology that stores byte-level stack information to generate ground truth, ensuring that task difficulty is preserved while maintaining universal applicability. This design enables fair evaluation across tasks while avoiding simplifications that could bias results. As a use case, we apply REBench to measure the reverse engineering performance of LLMs and the result demonstrates difficulties in complex tasks.

cs.CR↗
arXiv2026

Too Private to Tell: Practical Token Theft Attacks on Apple Intelligence

Apple Intelligence is a generative AI (GenAI) service provided by Apple on its devices. While offering a similar set of features as other similar GenAI services, Apple Intelligence is claimed to be designed with an extra focus on user security and privacy through a two-stage authentication and authorization design using anonymous access tokens. In this paper, we present our investigation into this token issuance mechanism with a goal to reveal possible vulnerabilities using traffic analysis, reverse engineering, and cross comparison with Apple's public documentation. Specifically, we present the Serpent attack, the first practical cross-device token replay attack against Apple Intelligence that allows the attacker to steal the access tokens from the victim's device and utilise them on a different device, with all usage rate-limited against the victim. We have achieved successful attacks on the latest macOS 26 Tahoe and demonstrated that an attacker, who even has used up its own allowance, can immediately regain access to Apple Intelligence service. We have responsibly disclosed the vulnerabilities to the vendors and received confirmation from Apple with CVE assigned and bounty given. Our results highlight a general lesson for built-in AI services: Anonymising identity does not by itself make the AI service secure; Enforcing non-transferability requires cryptographic binding to the rightful user.

cs.CR↗
arXiv2026

Styx: Collaborative and Private Data Processing With TEE-Enforced Sticky Policy

Protecting sensitive information in data-driven collaborations, such as AI training, while meeting the diverse requirements of multiple mutually distrusted stakeholders, is both crucial and challenging. This paper presents Styx, a novel framework to address this challenge by integrating sticky policies with Trusted Execution Environments (TEEs). At a high level, Styx employs a hardware-TEE-protected middleware with a programming language runtime to form a sandboxed environment for both the data processing and policy enforcement. We carefully designed a data processing workflow and pipelines to enable a strong yet flexible data-specific policy enforcement throughout the entire data lifecycle and data derivation to achieve data-in-use protection, data lifecycle protection and dynamic collaboration. We implemented Styx and demonstrated its ability to make collaborative computing, such as joint AI training, more secure, privacy-preserving, and policy-compliant. Our evaluation shows the performance overheads imposed by Styx are reasonable on single-node computation with the capability to scale to a large distributed multi-node deployment.

cs.CR↗
arXiv2026

Beyond OAuth: Task-Scoped Authorization for AI Agents via Natural Language Slices

AI agents increasingly execute users' natural-language (NL) tasks by calling Web services, yet today's Web authorizes these calls through OAuth, which grants permissions over operators (e.g., TRANSFER), not operations (operator plus operands, e.g., transfer $100 to Bob). This gap cannot be closed by refining scope granularity, because operands are combinatorial, quantitative, and often derived from runtime computations across servers. Operator-scoped authorization therefore inherently overprivileges agents. We propose Precise Task-Scoped Implicit Authorization (PAuth): submitting a concrete NL task implicitly authorizes exactly the operations its faithful execution requires, even when the agent is compromised (e.g., by malware or prompt injection). Each server independently derives an NL slice, a symbolic specification of the expected call inspired by program slicing, and server-produced values are wrapped in signed envelopes that bind concrete values to symbolic provenance. Together, they enforce that every operation, not just the operator, is consistent with the user's task, closing the gap that OAuth leaves open. We evaluate PAuth on AuthBench, a benchmark we build on top of AgentDojo and cross-validate on OpenClaw, spanning five service suites with 100 benign tasks and 634 adversarial calls. All 100 benign tasks are implicitly authorized and all 634 adversarial calls are blocked. Many tasks require multiple tool calls to complete. A unique value of PAuth is that it frees users from having to approve each call with concrete operand values, including intermediate results they never specified. This enhances both security and usability.

cs.CR↗
arXiv2026

A Survey of Reasoning in Autonomous Driving Systems: Open Challenges and Emerging Paradigms

The development of high-level autonomous driving (AD) is shifting from perception-centric limitations to a more fundamental bottleneck, namely, a deficit in robust and generalizable reasoning. Although current AD systems manage structured environments, they consistently falter in long-tail scenarios and complex social interactions that require human-like judgment. Meanwhile, the advent of large language and multimodal models (LLMs and MLLMs) presents a transformative opportunity to integrate a powerful cognitive engine into AD systems, moving beyond pattern matching toward genuine comprehension. However, a systematic framework to guide this integration is critically lacking. To bridge this gap, we provide a comprehensive review of this emerging field and argue that reasoning should be elevated from a modular component to the system's cognitive core. Specifically, we first propose a novel Cognitive Hierarchy to decompose the monolithic driving task according to its cognitive and interactive complexity. Building on this, we further derive and systematize seven core reasoning challenges, such as the responsiveness-reasoning trade-off and social-game reasoning. Furthermore, we conduct a dual-perspective review of the state-of-the-art, analyzing both system-centric approaches to architecting intelligent agents and evaluation-centric practices for their validation. Our analysis reveals a clear trend toward holistic and interpretable "glass-box" agents. In conclusion, we identify a fundamental and unresolved tension between the high-latency, deliberative nature of LLM-based reasoning and the millisecond-scale, safety-critical demands of vehicle control. For future work, a primary objective is to bridge the symbolic-to-physical gap by developing verifiable neuro-symbolic architectures, robust reasoning under uncertainty, and scalable models for implicit social negotiation.

cs.AI↗
arXiv2026

Atomicity for Agents: Exposing, Exploiting, and Mitigating TOCTOU Vulnerabilities in Browser-Use Agents

Browser-use agents are widely used for everyday tasks. They enable automated interaction with web pages through structured DOM based interfaces or vision language models operating on page screenshots. However, web pages often change between planning and execution, causing agents to execute actions based on stale assumptions. We view this temporal mismatch as a time of check to time of use (TOCTOU) vulnerability in browser-use agents. Dynamic or adversarial web content can exploit this window to induce unintended actions. We present a large scale empirical study of TOCTOU vulnerabilities in browser-use agents using a benchmark that spans synthesized and real world websites. Using this benchmark, we evaluate 10 popular open source agents and show that TOCTOU vulnerabilities are widespread. We design a lightweight mitigation based on pre-execution validation. It monitors DOM and layout changes during planning and validates the page state immediately before action execution. This approach reduces the risk of insecure execution and mitigates unintended side effects in browser-use agents.

cs.CR↗
arXiv2026

Web Agents Should Use Typed Actions Instead of Click-Based Browsing

This position paper argues that building a reliable agentic Web requires shifting from low-level interaction primitives to typed actions supported by a semantic layer. Today's web agents primarily operate through clicks, keystrokes, and DOM manipulation, which leads to brittle long-horizon behavior, high execution cost, and limited auditability. We propose web verbs as a concrete design for this layer. A verb exposes a web operation as a typed function with structured inputs, structured outputs, and documented behavior, whether it is backed by a server-side Web API or a maintained client-side workflow. Verb calls can carry preconditions, postconditions, policy tags, and logging hooks, allowing agents to synthesize concise programs with explicit control flow and data flow and to produce checkable execution traces. Using representative case studies, we illustrate how verb-level composition can produce correct, reproducible outcomes, while browser agents using low-level interaction primitives may produce brittle behavior or incorrect reasoning. We conclude with a call to action on standardization, developer tooling, and community processes needed to make this semantic layer deployable and trustworthy at web scale.

cs.AI↗
arXiv2026

SIN-Bench: Tracing Native Evidence Chains in Long-Context Multimodal Scientific Interleaved Literature

Evaluating whether multimodal large language models truly understand long-form scientific papers remains challenging: answer-only metrics and synthetic "Needle-In-A-Haystack" tests often reward answer matching without requiring a causal, evidence-linked reasoning trace in the document. We propose the "Fish-in-the-Ocean" (FITO) paradigm, which requires models to construct explicit cross-modal evidence chains within native scientific documents. To operationalize FITO, we build SIN-Data, a scientific interleaved corpus that preserves the native interleaving of text and figures. On top of it, we construct SIN-Bench with four progressive tasks covering evidence discovery (SIN-Find), hypothesis verification (SIN-Verify), grounded QA (SIN-QA), and evidence-anchored synthesis (SIN-Summary). We further introduce "No Evidence, No Score", scoring predictions when grounded to verifiable anchors and diagnosing evidence quality via matching, relevance, and logic. Experiments on eight MLLMs show that grounding is the primary bottleneck: Gemini-3-pro achieves the best average overall score (0.573), while GPT-5 attains the highest SIN-QA answer accuracy (0.767) but underperforms on evidence-aligned overall scores, exposing a gap between correctness and traceable support.

cs.CL↗
arXiv2025

A-MemGuard: A Proactive Defense Framework for LLM-Based Agent Memory

Large Language Model (LLM) agents use memory to learn from past interactions, enabling autonomous planning and decision-making in complex environments. However, this reliance on memory introduces a critical security risk: an adversary can inject seemingly harmless records into an agent's memory to manipulate its future behavior. This vulnerability is characterized by two core aspects: First, the malicious effect of injected records is only activated within a specific context, making them hard to detect when individual memory entries are audited in isolation. Second, once triggered, the manipulation can initiate a self-reinforcing error cycle: the corrupted outcome is stored as precedent, which not only amplifies the initial error but also progressively lowers the threshold for similar attacks in the future. To address these challenges, we introduce A-MemGuard (Agent-Memory Guard), the first proactive defense framework for LLM agent memory. The core idea of our work is the insight that memory itself must become both self-checking and self-correcting. Without modifying the agent's core architecture, A-MemGuard combines two mechanisms: (1) consensus-based validation, which detects anomalies by comparing reasoning paths derived from multiple related memories and (2) a dual-memory structure, where detected failures are distilled into ``lessons'' stored separately and consulted before future actions, breaking error cycles and enabling adaptation. Comprehensive evaluations on multiple benchmarks show that A-MemGuard effectively cuts attack success rates by over 95% while incurring a minimal utility cost. This work shifts LLM memory security from static filtering to a proactive, experience-driven model where defenses strengthen over time. Our code is available in https://github.com/TangciuYueng/AMemGuard

cs.CR↗
arXiv2025

MobiLLM: An Agentic AI Framework for Closed-Loop Threat Mitigation in 6G Open RANs

The evolution toward 6G networks is being accelerated by the Open Radio Access Network (O-RAN) paradigm -- an open, interoperable architecture that enables intelligent, modular applications across public telecom and private enterprise domains. While this openness creates unprecedented opportunities for innovation, it also expands the attack surface, demanding resilient, low-cost, and autonomous security solutions. Legacy defenses remain largely reactive, labor-intensive, and inadequate for the scale and complexity of next-generation systems. Current O-RAN applications focus mainly on network optimization or passive threat detection, with limited capability for closed-loop, automated response. To address this critical gap, we present an agentic AI framework for fully automated, end-to-end threat mitigation in 6G O-RAN environments. MobiLLM orchestrates security workflows through a modular multi-agent system powered by Large Language Models (LLMs). The framework features a Threat Analysis Agent for real-time data triage, a Threat Classification Agent that uses Retrieval-Augmented Generation (RAG) to map anomalies to specific countermeasures, and a Threat Response Agent that safely operationalizes mitigation actions via O-RAN control interfaces. Grounded in trusted knowledge bases such as the MITRE FiGHT framework and 3GPP specifications, and equipped with robust safety guardrails, MobiLLM provides a blueprint for trustworthy AI-driven network security. Initial evaluations demonstrate that MobiLLM can effectively identify and orchestrate complex mitigation strategies, significantly reducing response latency and showcasing the feasibility of autonomous security operations in 6G.

cs.CR↗
arXiv2025

AUTOVR: Automated UI Exploration for Detecting Sensitive Data Flow Exposures in Virtual Reality Apps

The rise of Virtual Reality (VR) has provided developers with an unprecedented platform for creating games and applications (apps) that require distinct inputs, different from those of conventional devices like smartphones. The Meta Quest VR platform, driven by Meta, has democratized VR app publishing and attracted millions of users worldwide. However, as the number of published apps grows, there is a notable lack of robust headless tools for user interface (UI) exploration and user event testing. To address this need, we present AUTOVR, an automatic framework for dynamic UI and user event interaction in VR apps built on the Unity Engine. Unlike conventional Android and GUI testers, AUTOVR analyzes the app's internal binary to reveal hidden events, resolves generative event dependencies, and utilizes them for comprehensive exploration of VR apps. Using sensitive data exposure as a performance metric, we compare AUTOVR with Android Monkey, a widely used headless Android GUI stress testing tool. Our empirical evaluation demonstrates AUTOVR's superior performance, triggering an order of magnitude of more sensitive data exposures and significantly enhancing the privacy of VR apps.

cs.CR↗