Papers

While the size of a data breach is typically measured by the number of (consumer, customer, or user) records exposed or compromised, its economic impact is generally measured from the point of view of the corporation suffering the data breach: cost in crisis management, legal fees, drop in stock price, and so on. This study examines whether it is possible to estimate the true cost, or the social cost of a data breach, measured by the impact on its victims and their out of pocket costs. To accomplish this we establish: (1) the estimation of the average direct financial losses of an identity theft (IDT) victim, including the opportunity cost of lost time, and healthcare expenditures associated with distress associated with identity theft; and (2) the estimation of increases in incidents of IDT that can be attributed to a major breach event. Our findings show that the average social cost per victim has declined significantly since 2016. Furthermore, we find that there is indeed a statistically significant increase in the number of IDTs following a mega-breach event when accounting for a discovery lag of 1-2 months post-breach. Applying our model to real-world cases allows us to estimate an upper and lower bound social cost of specific mega-breach events. We find that for the 2009 Heartland and 2013 Target breaches, even the conservative lower bound social cost estimate exceeded settlements by factors of 5 and 16, respectively. In contrast, the 2017 Equifax breach resulted in a lower bound estimate of $263.8 million, falling well within its $700 million settlement cap. While the Equifax upper bound estimate of $1.72 billion in social cost more than doubles this settlement, the narrowing gap between institutional liability and an incident's social cost provides empirical evidence of a market saturation effect that reduces the marginal damage of individual compromised records over time.

Host-acting agents promise a convenient interaction model in which users specify goals and the system determines how to realize them. We argue that this convenience introduces a distinct security problem: semantic under-specification in goal specification. User instructions are typically goal-oriented, yet they often leave process constraints, safety boundaries, persistence, and exposure insufficiently specified. As a result, the agent must complete missing execution semantics before acting, and this completion can produce risky host-side plans even when the user-stated goal is benign. In this paper, we develop a semantic threat model, present a taxonomy of semantic-induced risky completion patterns, and study the phenomenon through an OpenClaw-centered case study and execution-trace analysis. We further derive defense design principles for making execution boundaries explicit and constraining risky completion. These findings suggest that securing host-acting agents requires governing not only which actions are allowed at execution time, but also how goal-only instructions are translated into executable plans.

Open Radio Access Network (O-RAN) is a major advancement in the telecommunications field, providing standardized interfaces that promote interoperability between different vendors' technologies, thereby enhancing network flexibility and reducing operational expenses. By leveraging cutting-edge developments in network virtualization and artificial intelligence, O-RAN enhances operational efficiency and stimulates innovation within an open ecosystem. In the context of 6G, the potential capabilities of O-RAN have been significantly expanded, enabling ultra-reliable low-latency communication, terabit-level data rates, and seamless integration of terrestrial and non-terrestrial networks. Despite these benefits, its open architecture paradigm also brings critical security and privacy challenges, which, if not addressed, could compromise network integrity and data confidentiality. This paper conducts a comprehensive investigation into the security vulnerabilities and privacy issues associated with the O-RAN architecture in the context of the evolving 6G landscape, systematically categorizing fundamental vulnerabilities, meticulously examining potential attack vectors, and assessing current and future threats. In addition, this study also examines the existing and emerging security mechanisms of O-RAN and reviews the ongoing standardization activities aimed at strengthening the O-RAN security framework.

Multi-agent discussions have been widely adopted, motivating growing efforts to develop attacks that expose their vulnerabilities. In this work, we study a practical yet largely unexplored attack scenario, the discussion-monitored scenario, where anomaly detectors continuously monitor inter-agent communications and block detected adversarial messages. Although existing attacks are effective without discussion monitoring, we show that they exhibit detectable patterns and largely fail under such monitoring constraints. But does this imply that monitoring alone is sufficient to secure multi-agent discussions? To answer this question, we develop a novel attack method explicitly tailored to the discussion-monitored scenario. Extensive experiments demonstrate that effective attacks remain possible even under continuous monitoring, indicating that monitoring alone does not eliminate adversarial risks.

Tor is a low-latency anonymous communication network that protects user privacy by encrypting website traffic. However, recent website fingerprinting (WF) attacks have shown that encrypted traffic can still leak users' visited websites by exploiting statistical features such as packet size, direction, and inter-arrival time. Most existing WF attacks formulate the problem as a single-tab classification task, which significantly limits their effectiveness in realistic browsing scenarios where users access multiple websites concurrently, resulting in mixed traffic traces. To this end, we propose PrismWF, a multi-granularity patch-based Transformer for multi-tab WF attack. Specifically, we design a robust traffic feature representation for raw web traffic traces and extract multi-granularity features using convolutional kernels with different receptive fields. To effectively integrate information across temporal scales, the proposed model refines features through three hierarchical interaction mechanisms: inter-granularity detail supplementation from fine to coarse granularities, intra-granularity patch interaction with dedicated router tokens, and router-guided dual-level intra- and cross-granularity fusion. This design aligns with the cognitive logic of global coarse-grained reconnaissance and local fine-grained querying, enabling effective modeling of mixed traffic patterns in WF attack scenarios. Extensive experiments on various datasets and WF defenses demonstrate that our method achieves state-of-the-art performance compared to existing baselines.

Smart contracts have transformed decentralized finance, but flaws in their logic still create major security threats. Most existing vulnerability detection techniques focus on well-supported languages like Solidity, while low-resource counterparts such as Vyper remain largely underexplored due to scarce analysis tools and limited labeled datasets. Training a robust detection model directly on Vyper is particularly challenging, as collecting sufficiently large and diverse Vyper training datasets is difficult in practice. To address this gap, we introduce Sol2Vy, a novel framework that enables cross-language knowledge transfer from Solidity to Vyper, allowing vulnerability detection on Vyper using models trained exclusively on Solidity. This approach eliminates the need for extensive labeled Vyper datasets typically required to build a robust vulnerability detection model. We implement and evaluate Sol2Vy on various critical vulnerability types, including reentrancy, weak randomness, and unchecked transfer. Experimental results show that Sol2Vy, despite being trained exclusively on Solidity, achieves strong detection performance on Vyper contracts and significantly outperforms prior state-of-the-art methods.

With the rapid evolution of Large Language Model (LLM) agent ecosystems, centralized skill marketplaces have emerged as pivotal infrastructure for augmenting agent capabilities. However, these marketplaces face unprecedented security challenges, primarily stemming from semantic-behavioral inconsistency and inter-skill combinatorial risks, where individually benign skills induce malicious behaviors during collaborative invocation. To address these vulnerabilities, we propose SkillProbe, a multi-stage security auditing framework driven by multi-agent collaboration. SkillProbe introduces a "Skills-for-Skills" design paradigm, encapsulating auditing processes into standardized skill modules to drive specialized agents through a rigorous pipeline, including admission filtering, semantic-behavioral alignment detection, and combinatorial risk simulation. We conducted a large-scale evaluation using 8 mainstream LLM series across 2,500 real-world skills from ClawHub. Our results reveal a striking popularity-security paradox, where download volume is not a reliable proxy for security quality, as over 90% of high-popularity skills failed to pass rigorous auditing. Crucially, we discovered that high-risk skills form a single giant connected component within the risk-link dimension, demonstrating that cascaded risks are systemic rather than isolated occurrences. We hope that SkillProbe will inspire researchers to provide a scalable governance infrastructure for constructing a trustworthy Agentic Web. SkillProbe is accessible for public experience at skillhub.holosai.io.

While prior red-teaming efforts have focused on eliciting harmful text outputs from large language models (LLMs), such approaches fail to capture agent-specific vulnerabilities that emerge through multi-step tool execution, particularly in rapidly growing ecosystems such as the Model Context Protocol (MCP). To address this gap, we propose a trajectory-aware evolutionary search method, T-MAP, which leverages execution trajectories to guide the discovery of adversarial prompts. Our approach enables the automatic generation of attacks that not only bypass safety guardrails but also reliably realize harmful objectives through actual tool interactions. Empirical evaluations across diverse MCP environments demonstrate that T-MAP substantially outperforms baselines in attack realization rate (ARR) and remains effective against frontier models, including GPT-5.2, Gemini-3-Pro, Qwen3.5, and GLM-5, thereby revealing previously underexplored vulnerabilities in autonomous LLM agents. Code is available at https://github.com/pwnhyo/T-MAP.

Unmanned Aerial Vehicles (UAVs) are valuable for mission-critical systems like surveillance, rescue, or delivery. Not surprisingly, such systems attract cyberattacks, including Denial-of-Service (DoS) attacks to overwhelm the resources of mission drones (MDs). How can we defend UAV mission systems against DoS attacks? We adopt cyber deception as a defense strategy, in which honey drones (HDs) are proposed to bait and divert attacks. The attack and deceptive defense hinge upon radio signal strength: The attacker selects victim MDs based on their signals, and HDs attract the attacker from afar by emitting stronger signals, despite this reducing battery life. We formulate an optimization problem for the attacker and defender to identify their respective strategies for maximizing mission performance while minimizing energy consumption. To address this problem, we propose a novel approach, called HT-DRL. HT-DRL identifies optimal solutions without a long learning convergence time by taking the solutions of hypergame theory into the neural network of deep reinforcement learning. This achieves a systematic way to intelligently deceive attackers. We analyze the performance of diverse defense mechanisms under different attack strategies. Further, the HT-DRL-based HD approach outperforms existing non-HD counterparts up to two times better in mission performance while incurring low energy consumption.

The exact composition of mechanisms for which two differential privacy (DP) constraints hold simultaneously is studied. The resulting privacy region admits an exact representation as a mixture over compositions of mechanisms of heterogeneous DP guarantees, yielding a framework that naturally generalizes to the composition of mechanisms for which any number of DP constraints hold. This result is shown through a structural lemma for mixtures of binary hypothesis tests. Lastly, the developed methodology is applied to approximate $f$-DP composition.

AI agents today have passwords but no permission slips. They execute tool calls (fund transfers, database queries, shell commands, sub-agent delegation) with no standard mechanism to enforce authorization before the action executes. Current safety architectures rely on model alignment (probabilistic, training-time) and post-hoc evaluation (retrospective, batch). Neither provides deterministic, policy-based enforcement at the individual tool call level. We characterize this gap as the pre-action authorization problem and present the Open Agent Passport (OAP), an open specification and reference implementation that intercepts tool calls synchronously before execution, evaluates them against a declarative policy, and produces a cryptographically signed audit record. OAP enforces authorization decisions in a measured median of 53 ms (N=1,000). In a live adversarial testbed (4,437 authorization decisions across 1,151 sessions, $5,000 bounty), social engineering succeeded against the model 74.6% of the time under a permissive policy; under a restrictive OAP policy, a comparable population of attackers achieved a 0% success rate across 879 attempts. We distinguish pre-action authorization from sandboxed execution (contains blast radius but does not prevent unauthorized actions) and model-based screening (probabilistic), and show they are complementary. The same infrastructure that enforces security constraints (spending limits, capability scoping) also enforces quality gates, operational contracts, and compliance controls. The specification is released under Apache 2.0 (DOI: 10.5281/zenodo.18901596).

This paper presents a symmetric stream cipher that utilizes the dynamic properties of random cubic mappings in the complex plane to generate pseudo-random key streams. The system is based on the iterations of the random cubic polynomial $f_n(z)=z^3+c_n z$, where the parameters $c_n$ are chosen randomly from a disc of radius $\delta$ and with center at the origin, aiming to improve the chaotic behaviour and, consequently, the randomness of the generated sequence. The stability of the Julia set under small parameter perturbations, when $\delta < \delta_0\simeq 0.89$, is considered to ensure key consistency in noisy environments, such as 5G networks. On the other hand, for $\delta > 3$, the system exhibits instability and chaos, ideal for generating ultra-secure keys. The Python implementation integrates secure key derivation, robust key stream generation via warmed-up iteration, and an authenticated encryption scheme using the modern cryptographic primitives (HKDF andHMAC-SHA-256), to ensure message integrity and authenticity. Statistical analyses, including chi-square test and entropy calculation, are performed on the output of the key stream generator to evaluate its randomness and distribution. In addition, a complete statistical validation, compliant with NIST SP 800-22 standards in modern cryptography, was performed to enhance the proposed system's credibility.

Large Language Model (LLM) agents combine the chat interaction capabilities of LLMs with the power to interact with external tools and APIs. This enables them to perform complex tasks and act autonomously to achieve user goals. However, current agent systems operate on an all-or-nothing basis: an agent either has full access to an API's capabilities and a web page's content, or it has no access at all. This coarse-grained approach forces users to trust agents with more capabilities than they actually need for a given task. In this paper, we introduce AC4A, an access control framework for agents. As agents become more capable and autonomous, users need a way to limit what APIs or portions of web pages these agents can access, eliminating the need to trust them with everything an API or web page allows. Our goal with AC4A is to provide a framework for defining permissions that lets agents access only the resources they are authorized to access. AC4A works across both API-based and browser-based agents. It does not prescribe what permissions should be, but offers a flexible way to define and enforce them, making it practical for real-world systems. AC4A works by creating permissions granting access to resources, drawing inspiration from established access control frameworks like the one for the Unix file system. Applications define their resources as hierarchies and provide a way to compute the necessary permissions at runtime needed for successful resource access. We demonstrate the usefulness of AC4A in enforcing permissions over real-world APIs and web pages through case studies. The source code of AC4A is available at https://github.com/reSHARMA/AC4A

Sensor technologies have evolved to a point where it is now practical to monitor products along the supply chain. The collected data can be stored in a decentralized way using blockchain technology. However, ensuring the reliability of the sensed data is a critical challenge. In other words, we need to trust the data that we write to the blockchain. In this work, we propose ChainGuards, a decentralized system that uses product-specific rules to verify data collected across the supply chain, with particular focus on sensor-derived information, issuing warnings and triggering audits when anomalies are detected. We evaluated ChainGuards using data from a real cherry supply chain deployment. The result shows that the implemented solution provides reliable verification of supply chain data with low performance overhead, able to correctly detect data discrepancies and inconsistencies.

Graph neural network (GNN) is a powerful tool for analyzing graph-structured data. However, their vulnerability to adversarial attacks raises serious concerns, especially when dealing with sensitive information. Local Differential Privacy (LDP) offers a privacy-preserving framework for training GNNs, but its impact on adversarial robustness remains underexplored. This paper investigates adversarial attacks on LDP-protected GNNs. We explore how the privacy guarantees of LDP can be leveraged or hindered by adversarial perturbations. The effectiveness of existing attack methods on LDP-protected GNNs are analyzed and potential challenges in crafting adversarial examples under LDP constraints are discussed. Additionally, we suggest directions for defending LDP-protected GNNs against adversarial attacks. This work investigates the interplay between privacy and security in graph learning, highlighting the need for robust and privacy-preserving GNN architectures.

We propose a frequency-division multiplexed (FDM) continuous-variable quantum key distribution (CV-QKD) system with enhanced spectral efficiency through optimized channel spacing of low-symbol-rate signals. A four-channel 10-Mbaud FDM-CV-QKD system was experimentally demonstrated using Gaussian modulation, a transmitted local oscillator, and homodyne detection. Despite the inter-channel interference, under a finite-size scenario (m=1.25x10^6), the system achieved a 3.6-fold back-to-back secret key rate gain and outperformed the single-channel frequency-upconverted signal up to 26.8 km.

Large Language Models (LLMs) are increasingly adopted for vulnerability detection, yet their reasoning remains fundamentally unsound. We identify a root cause shared by both major mitigation paradigms (agent-based debate and retrieval augmentation): reasoning in an ungrounded deliberative space that lacks a bounded, hypothesis-specific evidence base. Without such grounding, agents fabricate cross-function dependencies, and retrieval heuristics supply generic knowledge decoupled from the repository's data-flow topology. Consequently, the resulting conclusions are driven by rhetorical persuasiveness rather than verifiable facts. To ground this deliberation, we present AEGIS, a novel multi-agent framework that shifts detection from ungrounded speculation to forensic verification over a closed factual substrate. Guided by a "From Clue to Verdict" philosophy, AEGIS first identifies suspicious code anomalies (clues), then dynamically reconstructs per-variable dependency chains for each clue via on-demand slicing over a repository-level Code Property Graph. Within this closed evidence boundary, a Verifier Agent constructs competing dialectical arguments for and against exploitability, while an independent Audit Agent scrutinizes every claim against the trace, exercising veto power to prevent hallucinated verdicts. Evaluation on the rigorous PrimeVul dataset demonstrates that AEGIS establishes a new state-of-the-art, achieving 122 Pair-wise Correct Predictions. To our knowledge, this is the first approach to surpass 100 on this benchmark. It reduces the false positive rate by up to 54.40% compared to leading baselines, at an average cost of $0.09 per sample without any task-specific training.

LLM agent frameworks increasingly offer checkpoint-restore for error recovery and exploration, advising developers to make external tool calls safe to retry. This advice assumes that a retried call will be identical to the original, an assumption that holds for traditional programs but fails for LLM agents, which re-synthesize subtly different requests after restore. Servers treat these re-generated requests as new, enabling duplicate payments, unauthorized reuse of consumed credentials, and other irreversible side effects; we term these semantic rollback attacks. We identify two attack classes, Action Replay and Authority Resurrection, validate them in a proof of concept experiment, and confirm that the problem has been independently acknowledged by framework maintainers. We propose ACRFence, a framework-agnostic mitigation that records irreversible tool effects and enforces replay-or-fork semantics upon restoration

Federated learning (FL) has attracted substantial attention in both academia and industry, yet its practical security posture remains poorly understood. In particular, a large body of poisoning research is evaluated under idealized assumptions about attacker participation, client homogeneity, and success metrics, which can substantially distort how security risks are perceived in deployed FL systems. This paper revisits FL security from a measurement perspective. We systematize three major sources of mismatch between research and practice: unrealistic poisoning threat models, the omission of hybrid heterogeneity, and incomplete metrics that overemphasize peak attack success while ignoring stability and utility cost. To study these gaps, we build TFLlib, a uniform evaluation framework that supports image, text, and tabular FL tasks and re-implements representative poisoning attacks under practical settings. Our empirical study shows that idealized evaluation often overstates security risk. Under practical settings, attack performance becomes markedly more dataset-dependent and unstable, and several attacks that appear consistently strong in idealized FL lose effectiveness or incur clear benign-task degradation once practical constraints are enforced. These findings further show that final-round attack success alone is insufficient for security assessment; practical measurement must jointly consider effectiveness, temporal stability, and collateral utility loss. Overall, this work argues that many conclusions in the FL poisoning literature are not directly transferable to real deployments. By tightening the threat model and using measurement protocols aligned with practice, we provide a more realistic view of the security risks faced by contemporary FL systems and distill concrete guidance for future FL security evaluation. Our code is available at https://github.com/xaddwell/TFLlib

Our analysis of recent Internet traces shows that up to 71% of flows contain suspicious behaviors indicative of low-volume network attacks such as port scans. However, distinguishing anomalous traffic in real time is challenging as each attack flow may comprise only a few packets. We extend prior work that tracks heavy hitter flows to also detect low-volume and slow attacks by combining the capabilities of both switches and SmartNICs. We flip the usual design approach by proposing an efficient filter data structure used to quickly route traffic marked as benign towards destination end-systems. We make careful use of limited programmable switch memory and pipeline stages, and complement them with SmartNIC resources to analyze the remaining traffic that may be anomalous. Using machine learning classifiers and intrusion detection rules deployed on the SmartNIC, we identify malicious source IPs, which then undergo more detailed forensics for attack mitigation. Finally, we develop a dataplane based protocol to rapidly coordinate data structure updates between these devices. We implement immUNITY in a testbed with Tofino v1 switch and Bluefield 3 SmartNIC, demonstrating its high accuracy, while minimizing traffic that's analyzed outside the switch.

Text-to-image models are commercially valuable assets often distributed under restrictive licenses, but such licenses are enforceable only when violations can be detected. Existing methods require pre-deployment watermarking or internal model access, which are unavailable in commercial API deployments. We present Compositional Semantic Fingerprinting (CSF), the first black-box method for attributing fine-tuned text-to-image models to protected lineages using only query access. CSF treats models as semantic category generators and probes them with compositional underspecified prompts that remain rare under fine-tuning. This gives IP owners an asymmetric advantage: new prompt compositions can be generated after deployment, while attackers must anticipate and suppress a much broader space of fingerprints. Across 6 model families (FLUX, Kandinsky, SD1.5/2.1/3.0/XL) and 13 fine-tuned variants, our Bayesian attribution framework enables controlled-risk lineage decisions, with all variants satisfying the dominance criterion.

We develop a formal theory of throughput in finite serial pipeline systems subject to stage multiplicative capacity perturbations, motivated by the deployment of AI tools in cybersecurity operations. A pipeline is a finite totally ordered set of stages each with a positive capacity throughput is the minimum stage capacity. An admissible multiplier assigns to each stage an improvement factor of at least one. We prove five theorems and one proposition. Theorems 1-2 give exact necessary and sufficient conditions. Throughput is unchanged if and only if at least one bottleneck retains multiplier 1, and throughput strictly increases if and only if every bottleneck has multiplier strictly greater than 1. Theorem 3 establishes that when a nonempty subset of stages is constrained to multiplier 1 the human authority constraint, throughput is bounded above by the smallest capacity among those stages, and this bound is tight under unbounded non human acceleration. Theorem 4 proves that in a pair of independent attacker defender pipelines, the attacker defender throughput ratio worsens for the defender if and only if the attacker relative throughput gain exceeds the defender. Theorem 5 proves that under a fixed false positive fraction model, useful throughput is constant not decreasing above the investigation capacity, establishing that a commonly asserted paradoxical decline is impossible in that model. Proposition 6 shows that replacing the fixed fraction with a rate dependent precision function that is strictly decreasing suffices to recover the intended decline. All proofs are elementary, using only finite minima, real number order properties, and pointwise multiplicative structure.

Modern cloud inference creates a two sided privacy problem where users reveal sensitive inputs to providers, while providers must execute proprietary model weights inside potentially leaky execution environments. Fully homomorphic encryption (FHE) offers cryptographic guarantees but remains prohibitively expensive for modern architectures. We argue that progress requires co-design where specializing FHE schemes/compilers for the static structure of inference circuits, while simultaneously constraining inference architectures to reduce dominant homomorphic cost drivers. We outline a meet in the middle agenda and concrete optimization targets on both axes.

We present Hawkeye, a system for analyzing and reproducing GPU-level arithmetic operations. Using our framework, anyone can re-execute on a CPU the exact matrix multiplication operations underlying a machine learning model training or inference workflow that was executed on an NVIDIA GPU, without any precision loss. This is in stark contrast to prior approaches to verifiable machine learning, which either introduce significant computation overhead to the original model owner, or suffer from non-robustness and quality degradation. The main technical contribution of Hawkeye is a systematic sequence of carefully crafted tests that study rounding direction, subnormal number handling, and order of (non-associative) accumulation during matrix multiplication on NVIDIA's Tensor Cores. We test and evaluate our framework on multiple NVIDIA GPU architectures ( Ampere, Hopper, and Lovelace) and precision types (FP16, BFP16, FP8). In all test cases, Hawkeye enables perfect reproduction of matrix multiplication on a CPU, paving the way for efficient and trustworthy third-party auditing of ML model training and inference. We provide source code for Hawkeye at https://github.com/badasherez/gpu-simulator.

Memory poisoning attacks for Agentic AI and multi-agent systems (MAS) have recently caught attention. It is partially due to the fact that Large Language Models (LLMs) facilitate the construction and deployment of agents. Different memory systems are being used nowadays in this context, including semantic, episodic, and short-term memory. This distinction between the different types of memory systems focuses mostly on their duration but also on their origin and their localization. It ranges from the short-term memory originated at the user's end localized in the different agents to the long-term consolidated memory localized in well established knowledge databases. In this paper, we first present the main types of memory systems, we then discuss the feasibility of memory poisoning attacks in these different types of memory systems, and we propose mitigation strategies. We review the already existing security solutions to mitigate some of the alleged attacks, and we discuss adapted solutions based on cryptography. We propose to implement local inference based on private knowledge retrieval as an example of mitigation strategy for memory poisoning for semantic memory. We also emphasize actual risks in relation to interactions between agents, which can cause memory poisoning. These latter risks are not so much studied in the literature and are difficult to formalize and solve. Thus, we contribute to the construction of agents that are secure by design.

Agent frameworks increasingly encode tool-using behavior as explicit workflow graphs, yet safety enforcement remains a runtime concern. These frameworks expose analyzable graph structure through their APIs, enabling pre-deployment static verification of safety properties that runtime guardrails can only check reactively. This paper presents Agentproof, a system that automatically extracts a unified abstract graph model from four major agent frameworks (LangGraph, CrewAI, AutoGen, Google ADK), applies six structural checks with witness trace generation, and evaluates temporal safety policies via a DSL compiled to deterministic finite automata, both statically through a graph x DFA product construction and at runtime over event traces. Unlike general-purpose model checkers, Agentproof requires no manual modeling. In a curated benchmark of 18 author-constructed workflows, 27% of the benchmark contain structural defects (dead-end nodes, unreachable exits) and 55% violate a human-gate policy when enforced, distinct categories that prior work conflates. All 15 temporal policies defined fit within the seven-form DSL fragment, and verification completes in sub-second time for graphs up to 5,000 nodes. The corpus serves as a reproducible benchmark for evaluating static verification tools rather than as a prevalence study; defect rates reflect tool detection capability on a targeted benchmark, not base rates in production systems. Nonetheless, static graph verification complements runtime guardrails by catching topology-level defects that runtime tools miss unless the offending path is exercised.

Mobile advertising dominates app monetization but introduces risks ranging from intrusive user experience to malware delivery. Existing detection methods rely either on static analysis, which misses runtime behaviors, or on heuristic UI exploration, which struggles with sparse and obfuscated ads. In this paper, we present MANA, the first agentic multimodal reasoning framework for mobile ad detection. MANA integrates static, visual, temporal, and experiential signals into a reasoning-guided navigation strategy that determines not only how to traverse interfaces but also where to focus, enabling efficient and robust exploration. We implement and evaluate MANA on commercial smartphones over 200 apps, achieving state-of-the-art accuracy and efficiency. Compared to baselines, it improves detection accuracy by 30.5%-56.3% and reduces exploration steps by 29.7%-63.3%. Case studies further demonstrate its ability to uncover obfuscated and malicious ads, underscoring its practicality for mobile ad auditing and its potential for broader runtime UI analysis (e.g., permission abuse). Code and dataset are available at https://github.com/MANA-2026/MANA.

Low-level C programs remain highly vulnerable to out-of-bounds memory corruption. State-of-the-art precise defenses either introduce severe runtime overhead due to metadata memory lookups, or break standard C semantics by disallowing partial structs or the creation of an object's end address (EA), a legal operation ubiquitous in real-world C code. Conversely, practical alignment-based solutions achieve efficiency only by relaxing protected bounds. We present PRISM, a precise, zero-lookup object-bounds scheme that eliminates these restrictions. PRISM compresses a 47-bit EA into the 17-bit unused tag area of a 64-bit pointer. By enforcing the invariant that a statically known starting address (KSA) cannot exceed the EA, PRISM completely eliminates the need for costly metadata memory fetches in nearly all bounds checks, while strictly retaining precise object bounds. Our invariant also simplifies the lower-bound checks in existing alignment-based solutions, thus improving their performance. To achieve high throughput, PRISM introduces q-padding, an optimization that safely removes bounds checks for constant-offset accesses (such as struct fields) while maintaining precise, byte-level protection for the variable-indexed accesses primarily exploited by attackers. Evaluated on SPEC 2017, PRISM achieves an arithmetic mean CPU overhead of 46.1% with a 32-byte q-padding (dropping to 31.3% in a 32-bit address space). On highly concurrent, real-world workloads, PRISM secures a fully saturated Apache web server with only an 11.1% throughput reduction, demonstrating its readiness for production deployment. Furthermore, PRISM successfully detected an out-of-bounds violation in gcc that prior tools missed due to their lack of support for partial structs.

Modern graph learning systems often combine links with text, as in citation networks with abstracts or social graphs with user posts. In such systems, text is usually easier to edit than graph structure, which creates a practical security risk: an attacker may hide a small malicious cue in training text and later use it to trigger incorrect predictions. This paper studies that risk in a realistic setting where the attacker edits only node text and leaves the graph unchanged. We propose TAGBD, a graph-aware backdoor attack that first selects training nodes that are easier to manipulate, then generates stealthy poison text with a shadow graph model, and finally injects the text by replacing the original content or appending a short phrase. Experiments on three benchmark datasets show that TAGBD achieves very high attack success rates, transfers across different graph models, and remains effective under common defenses. These results show that inconspicuous poison text alone can serve as a reliable attack channel in text-attributed graphs, highlighting the need for defenses that inspect both node content and graph structure.

The use of ML in cybersecurity has long been impaired by generalization issues: Models that work well in controlled scenarios fail to maintain performance in production. The root cause often lies in ML algorithms learning superficial patterns (shortcuts) rather than underlying cybersecurity concepts. We investigate contrastive multi-modal learning as a first step towards improving ML performance in cybersecurity tasks. We aim at transferring knowledge from data-rich modalities, such as text, to data-scarce modalities, such as payloads. We set up a case study on threat classification and propose a two-stage multi-modal contrastive learning framework that uses textual vulnerability descriptions to guide payload classification. First, we construct a semantically meaningful embedding space using contrastive learning on descriptions. Then, we align payloads to this space, transferring knowledge from text to payloads. We evaluate the approach on a large-scale private dataset and a synthetic benchmark built from public CVE descriptions and LLM-generated payloads. The methodology appears to reduce shortcut learning over baselines on both benchmarks. We release our synthetic benchmark and source code as open source.