Papers

The Ring-Learning With Errors (RLWE) problem forms the backbone of highly efficient Fully Homomorphic Encryption (FHE) schemes. A significant component of the RLWE public key and ciphertext of the form $(b,a)$ is the uniformly random polynomial $a \in R_q$ . While essential for security, the communication overhead of transmitting $a$ from client to server, and inputting it into a hardware accelerator, can be substantial, especially for FHE accelerators aiming at high acceleration factors. A known technique in reducing this overhead generates $a$ from a small seed on the client side via a deterministic process, transmits only the seed, and generates $a$ on-the-fly within the accelerator. Challenges in the hardware implementation of an accelerator include wiring (density and power), compute area, compute power as well as flexibility in scheduling of on-the-fly generation instructions. This extended abstract proposes a concrete scheme and parameters wherein these practical challenges are addressed. We detail the benefits of our approach, which maintains the reduction in communication latency and memory footprint, while allowing parallel generation of uniformly distributed samples, relaxed wiring requirements, unrestricted randomaccess to RNS limbs, and results in an extremely low overhead on the client side (i.e. less than 3%) during the key generation process. The proposed scheme eliminates the need for thick metal layers for randomness distribution and prevents the power consumption of the PRNG subsystem from scaling prohibitively with the acceleration factor, potentially saving tens of Watts per accelerator chip in high-throughput configurations.

LLM-based code interpreter agents are increasingly deployed in critical workflows, yet their robustness against risks introduced by their code execution capabilities remains underexplored. Existing benchmarks are limited to static datasets or simulated environments, failing to capture the security risks arising from dynamic code execution, tool interactions, and multi-turn context. To bridge this gap, we introduce CIBER, an automated benchmark that combines dynamic attack generation, isolated secure sandboxing, and state-aware evaluation to systematically assess the vulnerability of code interpreter agents against four major types of adversarial attacks: Direct/Indirect Prompt Injection, Memory Poisoning, and Prompt-based Backdoor. We evaluate six foundation models across two representative code interpreter agents (OpenInterpreter and OpenCodeInterpreter), incorporating a controlled study of identical models. Our results reveal that Interpreter Architecture and Model Alignment Set the Security Baseline. Structural integration enables aligned specialized models to outperform generic SOTA models. Conversely, high intelligence paradoxically increases susceptibility to complex adversarial prompts due to stronger instruction adherence. Furthermore, we identify a "Natural Language Disguise" Phenomenon, where natural language functions as a significantly more effective input modality than explicit code snippets (+14.1% ASR), thereby bypassing syntax-based defenses. Finally, we expose an alarming Security Polarization, where agents exhibit robust defenses against explicit threats yet fail catastrophically against implicit semantic hazards, highlighting a fundamental blind spot in current pattern-matching protection approaches.

Age estimation systems are increasingly deployed as gatekeepers for age-restricted online content, yet their robustness to cosmetic modifications has not been systematically evaluated. We investigate whether simple, household-accessible cosmetic changes, including beards, grey hair, makeup, and simulated wrinkles, can cause AI age estimators to classify minors as adults. To study this threat at scale without ethical concerns, we simulate these physical attacks on 329 facial images of individuals aged 10 to 21 using a VLM image editor (Gemini 2.5 Flash Image). We then evaluate eight models from our prior benchmark: five specialized architectures (MiVOLO, Custom-Best, Herosan, MiViaLab, DEX) and three vision-language models (Gemini 3 Flash, Gemini 2.5 Flash, GPT-5-Nano). We introduce the Attack Conversion Rate (ACR), defined as the fraction of images predicted as minor at baseline that flip to adult after attack, a population-agnostic metric that does not depend on the ratio of minors to adults in the test set. Our results reveal that a synthetic beard alone achieves 28 to 69 percent ACR across all eight models; combining all four attacks shifts predicted age by +7.7 years on average across all 329 subjects and reaches up to 83 percent ACR; and vision-language models exhibit lower ACR (59 to 71 percent) than specialized models (63 to 83 percent) under the full attack, although the ACR ranges overlap and the difference is not statistically tested. These findings highlight a critical vulnerability in deployed age-verification pipelines and call for adversarial robustness evaluation as a mandatory criterion for model selection.

Autonomous AI agents can now programmatically hire human workers through marketplaces using REST APIs and Model Context Protocol (MCP) integrations. This creates an attack surface analogous to CAPTCHA-solving services but with physical-world reach. We present an empirical measurement study of this threat, analyzing 303 bounties from RENTAHUMAN.AI, a marketplace where agents post tasks and manage escrow payments. We find that 99 bounties (32.7%), originate from programmatic channels (API keys or MCP). Using a dual-coder methodology (ąppa} = 0.86 ), we identify six active abuse classes: credential fraud, identity impersonation, automated reconnaissance, social media manipulation, authentication circumvention, and referral fraud, all purchasable for a median of $25 per worker. A retrospective evaluation of seven content-screening rules flags 52 bounties (17.2%) with a single false positive, demonstrating that while basic defenses are feasible, they are currently absent.

Traditional database fuzzing techniques primarily focus on syntactic correctness and general SQL structures, leaving critical yet obscure DBMS features, such as system-level modes (e.g., GTID), programmatic constructs (e.g., PROCEDURE), advanced process commands (e.g., KILL), largely underexplored. Although rarely triggered by typical inputs, these features can lead to severe crashes or security issues when executed under edge-case conditions. In this paper, we present FuzzySQL, a novel LLM-powered adaptive fuzzing framework designed to uncover subtle vulnerabilities in DBMS special features. FuzzySQL combines grammar-guided SQL generation with logic-shifting progressive mutation, a novel technique that explores alternative control paths by negating conditions and restructuring execution logic, synthesizing structurally and semantically diverse test cases. To further ensure deeper execution coverage of the back end, FuzzySQL employs a hybrid error repair pipeline that unifies rule-based patching with LLM-driven semantic repair, enabling automatic correction of syntactic and context-sensitive failures. We evaluate FuzzySQL across multiple DBMSs, including MySQL, MariaDB, SQLite, PostgreSQL and Clickhouse, uncovering 64 vulnerabilities, 27 of which are tied to under-tested DBMS special features. As of this writing, 60 cases have been confirmed with 9 assigned CVE identifiers, 31 already fixed by vendors, and additional vulnerabilities scheduled to be patched in upcoming releases. Our results highlight the limitations of conventional fuzzers in semantic feature coverage and demonstrate the potential of LLM-based fuzzing to discover deeply hidden bugs in complex database systems.

Trusted Execution Environments (TEEs) (e.g., Intel SGX and ArmTrustZone) aim to protect sensitive computation from a compromised operating system, yet real deployments remain vulnerable to microarchitectural leakage, side-channel attacks, and fault injection. In parallel, security teams increasingly rely on Large Language Model (LLM) assistants as security advisors for TEE architecture review, mitigation planning, and vulnerability triage. This creates a socio-technical risk surface: assistants may hallucinate TEE mechanisms, overclaim guarantees (e.g., what attestation does and does not establish), or behave unsafely under adversarial prompting. We present a red-teaming study of two prevalently deployed LLM assistants in the role of TEE security advisors: ChatGPT-5.2 and Claude Opus-4.6, focusing on the inherent limitations and transferability of prompt-induced failures across LLMs. We introduce TEE-RedBench, a TEE-grounded evaluation methodology comprising (i) a TEE-specific threat model for LLM-mediated security work, (ii) a structured prompt suite spanning SGX and TrustZone architecture, attestation and key management, threat modeling, and non-operational mitigation guidance, along with policy-bound misuse probes, and (iii) an annotation rubric that jointly measures technical correctness, groundedness, uncertainty calibration, refusal quality, and safe helpfulness. We find that some failures are not purely idiosyncratic, transferring up to 12.02% across LLM assistants, and we connect these outcomes to secure architecture by outlining an "LLM-in-the-loop" evaluation pipeline: policy gating, retrieval grounding, structured templates, and lightweight verification checks that, when combined, reduce failures by 80.62%.

Modern organizations increasingly face cybersecurity incidents driven by human behaviour rather than technical failures. To address this, we propose a conceptual security framework that integrates a hybrid Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) model to analyze biometric and environmental data for context-aware security decisions. The CNN extracts spatial patterns from sensor data, while the LSTM captures temporal dynamics associated with human error susceptibility. The model achieves 84% accuracy, demonstrating its ability to reliably detect conditions that lead to elevated human-centred cyber risk. By enabling continuous monitoring and adaptive safeguards, the framework supports proactive interventions that reduce the likelihood of human-driven cyber incidents

The proliferation of autonomous AI agents capable of executing real-world actions - filesystem operations, API calls, database modifications, financial transactions - introduces a class of safety risk not addressed by existing content-moderation infrastructure. Current text-safety systems evaluate linguistic content for harm categories such as violence, hate speech, and sexual content; they are architecturally unsuitable for evaluating whether a proposed action falls within an agent's authorized operational scope. We present ILION (Intelligent Logic Identity Operations Network), a deterministic execution gate for agentic AI systems. ILION employs a five-component cascade architecture - Transient Identity Imprint (TII), Semantic Vector Reference Frame (SVRF), Identity Drift Control (IDC), Identity Resonance Score (IRS) and Consensus Veto Layer (CVL) - to classify proposed agent actions as BLOCK or ALLOW without statistical training or API dependencies. The system requires zero labeled data, operates in sub-millisecond latency, and produces fully interpretable verdicts. We evaluate ILION on ILION-Bench v2, a purpose-built benchmark of 380 test scenarios across eight attack categories with 39% hard-difficulty adversarial cases and a held-out development split. ILION achieves F1 = 0.8515, precision = 91.0%, and a false positive rate of 7.9% at a mean latency of 143 microseconds. Comparative evaluation against three baselines - Lakera Guard (F1 = 0.8087), OpenAI Moderation API (F1 = 0.1188), and Llama Guard 3 (F1 = 0.0105) - demonstrates that existing text-safety infrastructure systematically fails on agent execution safety tasks due to a fundamental task mismatch. ILION outperforms the best commercial baseline by 4.3 F1 points while operating 2,000 times faster with a false positive rate four times lower.

LLMs see the world as a single stream of text, partitioned into roles like <user> or <tool>. We trace prompt injection to role confusion: models perceive the source of text from how it sounds, not its labeled role. A command hidden in a webpage hijacks an agent simply because it sounds like <user> text, despite its <tool> label. We design role probes to measure how LLMs internally perceive "who is speaking," and find that injected text occupies the same representational space as the trusted role it imitates. We demonstrate this with CoT Forgery, a zero-shot attack that injects fabricated reasoning into user prompts and tool outputs. Models mistake the forgery for their own thoughts, yielding 60% attack success against frontier models with near-zero baselines. Strikingly, the degree of role confusion predicts attack success before a single token is generated. This mechanism generalizes beyond CoT Forgery to standard agent prompt injections, revealing prompt injection as a measurable consequence of role perception. To the model, sounding like a role is indistinguishable from being one. Project page and writeup: https://role-confusion.github.io

Security monitoring systems typically treat anomaly detection as identifying statistical deviations from observed data distributions. In cryptographic traffic analysis, however, violations are defined not by rarity but by explicit policy constraints, including key reuse prohibition, downgrade prevention, and bounded key lifetimes. This fundamental mismatch limits the interpretability and adaptability of conventional anomaly detection methods. We introduce INTACT (INTent-Aware Cryptographic Traffic), a policy-conditioned framework that reformulates violation detection as conditional constraint learning. Instead of learning a static decision boundary over behavioral features, INTACT models the probability of violation conditioned on both observed behavior and declared security intent. The architecture factorizes representation learning into behavioral and intent encoders whose fused embeddings produce a violation score, yielding a policy-parameterized family of decision boundaries. We evaluate the framework on a real-world network flow dataset and a 210,000-trace synthetic multi-intent cryptographic dataset. INTACT matches or exceeds strong unsupervised and supervised baselines, achieving near-perfect discrimination (AUROC up to 1.0000) in the real dataset and consistent superiority in detecting relational and composite violations in the synthetic setting. These results demonstrate that explicit intent conditioning improves discrimination, interpretability, and robustness in cryptographic monitoring.

The growing reliance on AI-identified digital evidence raises significant concerns about its reliability, particularly as large language models (LLMs) are increasingly integrated into forensic investigations. This paper proposes a structured framework that automates forensic artifact extraction, refines data through LLM-driven analysis, and validates results using a Digital Forensic Knowledge Graph (DFKG). Evaluated on a 13 GB forensic image dataset containing 61 applications, 2,864 databases, and 5,870 tables, the framework ensures artifact traceability and evidentiary consistency through deterministic Unique Identifiers (UIDs) and forensic cross-referencing. We propose this methodology to address challenges in ensuring the credibility and forensic integrity of AI-identified evidence, reducing classification errors, and advancing scalable, auditable methodologies. A comprehensive case study on this dataset demonstrates the framework's effectiveness, achieving over 95 percent accuracy in artifact extraction, strong support of chain-of-custody adherence, and robust contextual consistency in forensic relationships. Key results validate the framework's ability to enhance reliability, reduce errors, and establish a legally sound paradigm for AI-assisted digital forensics.

AI agents increasingly require direct, structured access to application data and actions, but production deployments still struggle to express and verify the governance properties that matter in practice: least-privilege authorization, controlled write execution, predictable failure handling, abuse resistance, and auditability. This paper introduces OpenPort Protocol (OPP), a governance-first specification for exposing application tools through a secure server-side gateway that is model- and runtime-neutral and can bind to existing tool ecosystems. OpenPort defines authorization-dependent discovery, stable response envelopes with machine-actionable agent.* reason codes, and an authorization model combining integration credentials, scoped permissions, and ABAC-style policy constraints. For write operations, OpenPort specifies a risk-gated lifecycle that defaults to draft creation and human review, supports time-bounded auto-execution under explicit policy, and enforces high-risk safeguards including preflight impact binding and idempotency. To address time-of-check/time-of-use drift in delayed approval flows, OpenPort also specifies an optional State Witness profile that revalidates execution-time preconditions and fails closed on state mismatch. Operationally, the protocol requires admission control (rate limits/quotas) with stable 429 semantics and structured audit events across allow/deny/fail paths so that client recovery and incident analysis are deterministic. We present a reference runtime and an executable governance toolchain (layered conformance profiles, negative security tests, fuzz/abuse regression, and release-gate scans) and evaluate the core profile at a pinned release tag using artifact-based, externally reproducible validation.

Managing personal health data is a challenge in today's fragmented and institution-centric healthcare ecosystem. Individuals often lack meaningful control over their medical records, which are scattered across incompatible systems and formats. This vision paper presents Health+, a user-centric, multimodal health data management system that empowers individuals (including those with limited technical expertise) to upload, query, and share their data across modalities (e.g., text, images, reports). Rather than aiming for institutional overhaul, Health+ emphasizes individual agency by providing intuitive interfaces and intelligent recommendations for data access and sharing. At the system level, it tackles the complexity of storing, integrating, and securing heterogeneous health records, ensuring both efficiency and privacy. By unifying multimodal data and prioritizing patients, Health+ lays the foundation for a more connected, interpretable, and user-controlled health information ecosystem.

Large language models (LLMs) acquire a large amount of knowledge through pre-training on vast and diverse corpora. While this endows LLMs with strong capabilities in generation and reasoning, it amplifies risks associated with sensitive, copyrighted, or harmful content in training data. LLM unlearning, which aims to remove specific knowledge encoded within models, is a promising technique to reduce these risks. However, existing LLM unlearning methods often force LLMs to generate random or incoherent answers due to their inability to alter the encoded knowledge precisely. To achieve effective unlearning at the knowledge level of LLMs, we propose Knowledge Unlearning by Deviating representAtion (KUDA). We first utilize causal tracing to locate specific layers for target knowledge storage. We then design a new unlearning objective that induces the model's representations to deviate from its original position in the phase of knowledge removal, thus disrupting the ability to associate with the target knowledge. To resolve the optimization conflicts between forgetting and retention, we employ a relaxation null-space projection mechanism to mitigate the disruption to the representation space of retaining knowledge. Extensive experiments on representative benchmarks, WMDP and MUSE, demonstrate that KUDA outperforms most existing baselines by effectively balancing knowledge removal and model utility retention. Source code is available at https://github.com/AceNagi/KUDA.

Risk matrices (heatmaps) are widely used for information and cyber risk management and decision-making, yet they are often too coarse for today's resilience-driven organizational and system landscapes. Likelihood and impact (the two dimensions represented in a heatmap) can vary with operational conditions, third-party dependencies, and the effectiveness of technical and organizational controls. At the same time, organizations cannot afford to analyze and operationalize every identified risk with equal depth using more sophisticated methods, telemetry, and real-time decision logic. We therefore propose a traceable triage pipeline that connects broad, context-sensitive screening with selective deep-dive analysis of material risks. The Hagenberg Risk Management Process presented in this paper integrates three steps: (i) context-aware prioritization using multidimensional polar heatmaps to compare risks across multiple operational states, (ii) Bowtie analysis for triaged risks to structure causes, consequences, and barriers, and (iii) an automated transformation of Bowties into directed acyclic graphs as the structural basis for Bayesian networks. A distinctive feature is the explicit representation of barriers as activation nodes in the resulting graph, making control points visible and preparing for later intervention and what-if analyses. The approach is demonstrated on an instant-payments gateway scenario in which a faulty production change under peak load leads to cascading degradation and transaction loss; DORA serves as the reference framework for resilience requirements. The result is an end-to-end, tool-supported workflow that improves transparency, auditability, and operational readiness from prioritization to monitoring-oriented models.

Non-fungible tokens (NFTs) on Ethereum currently follow a binary mobility paradigm: ERC-721 enables unrestricted transfers, whereas SBTs (ERC-5192) prohibit transfers entirely. We identify a design gap in which no standard mechanism supports bounded transferability, where ownership mobility is allowed but limited to a finite number of programmable transfers. We study counted NFT transfers and introduce ERC-7634 as a minimal realization compatible with ERC-721. The design augments each token with a transfer counter and configurable cap L, allowing ownership to evolve under a finite transfer budget. ERC-7634 defines a minimal extension interface with three lightweight functions (transferCountOf, setTransferLimit, and transferLimitOf), two events, and native-transfer hooks, requiring fewer than 60 additional lines of Solidity while preserving full backward compatibility with existing NFT infrastructure. We analyze behavioral and economic consequences of counted transfers. Our results reveal (i) a mobility premium induced by remaining transfer capacity, (ii) a protocol-level costing signal that can deter wash trading in cap-aware markets through irreversible budget consumption, (iii) bounded recursive collateralization enabled by limited ownership turnover, and (iv) associated security and gas-cost implications, including wrapper-bypass trade-offs. Evaluation on calibrated simulations shows that moderate limits (e.g., L = 10) affect fewer than 15% of tokens under representative transfer distributions, while repeated manipulation becomes unprofitable after a few cycles in a cap-aware pricing model; the additional gas overhead remains below 11% per transfer. We further position ERC-7634 within the NFT mobility design space, derive practical cap-selection guidelines, and discuss post-cap ownership outcomes including soulbound conversion, auto-burn, and provenance freeze.

Image-generative models are widely deployed across industries. Recent studies show that they can be exploited to produce unacceptable content. Existing mitigation strategies rely on prompt filtering and safety-aware training, both of which can be bypassed and often degrade generative quality. In this work, we propose ReVision, a training-free, prompt-based, post-hoc safety framework for image-generation pipeline. ReVision acts as a post-generation safeguard by analyzing generated images and selectively editing unsafe concepts without altering the underlying generator. Prior post-hoc editing methods often rely on imprecise spatial localization, limiting deployability, in multi-concept scenes. To address this limitation, ReVision introduces a VLM-assisted spatial gating mechanism for instance-consistent localization, enabling integrity-preserving edits. We introduce an 800-image benchmark spanning single- and multi-unsafe-concept images, each composed alongside benign concepts in shared scenes. On this benchmark, ReVision improves CLIP alignment toward safe prompts by +0.121, reduces multi-concept background LPIPS from 0.166 to 0.058, and eliminates NudeNet detections (70.51 -> 0). Across external benchmarks, ReVision outperforms prior methods, and a human study shows it reduces recognizability of unacceptable content from 96% to 10%.

Robust reversible watermarking (RRW) enables copyright protection for images while overcoming the limitation of distortion introduced by watermark itself. Current RRW schemes typically employ a two-stage framework, which fails to achieve simultaneous robustness and reversibility within a single watermarking, and functional interference between the two watermarks results in performance degradation in multiple terms such as capacity and imperceptibility. We propose SiGRRW, a single-watermark RRW framework, which is applicable to both generative models and natural images. We introduce a novel guiding strategy to generate guiding images, serving as the guidance for embedding and recovery. The watermark is reversibly embedded with the guiding residual, which can be calculated from both cover images and watermark images. The proposed framework can be deployed either as a plug-and-play watermarking layer at the output stage of generative models, or directly applied to natural images. Extensive experiments demonstrate that SiGRRW effectively enhances imperceptibility and robustness compared to existing RRW schemes while maintaining lossless recovery of cover images, with significantly higher capacity than conventional schemes.

The critical need for transparent and trustworthy machine learning in cybersecurity operations drives the development of this integrated Explainable AI (XAI) framework. Our methodology addresses three fundamental challenges in deploying AI for threat detection: handling massive datasets through Strategic Sampling Methodology that preserves class distributions while enabling efficient model development; ensuring experimental rigor via Automated Data Leakage Prevention that systematically identifies and removes contaminated features; and providing operational transparency through Integrated XAI Implementation using SHAP analysis for model-agnostic interpretability across algorithms. Applied to the CIC-IDS2017 dataset, our approach maintains detection efficacy while reducing computational overhead and delivering actionable explanations for security analysts. The framework demonstrates that explainability, computational efficiency, and experimental integrity can be simultaneously achieved, providing a robust foundation for deploying trustworthy AI systems in security operations centers where decision transparency is paramount.

Mixture-of-Experts (MoE) offers flexible graph reasoning by combining multiple views of a graph through a learned router. We investigate routing-aware explanations for MoE graph models in malware detection using control flow graphs (CFGs). Our architecture builds diversity at two levels. At the node level, each layer computes multiple neighborhood statistics and fuses them with an MLP, guided by a degree reweighting factor rho and a pooling choice lambda in {mean, std, max}, producing distinct node representations that capture complementary structural cues in CFGs. At the readout level, six experts, each tied to a specific (rho, lambda) view, output graph-level logits that the router weights into a final prediction. Post-hoc explanations are generated with edge-level attributions per expert and aggregated using the router gates so the rationale reflects both what each expert highlights and how strongly it is selected. Evaluated against single-expert GNN baselines such as GCN, GIN, and GAT on the same CFG dataset, the proposed MoE achieves strong detection accuracy while yielding stable, faithful attributions under sparsity-based perturbations. The results indicate that making the router explicit and combining multi-statistic node encoding with expert-level diversity can improve the transparency of MoE decisions for malware analysis.

As AI shifts from human-in-the-loop interfaces to autonomous multi-agent systems capable of real-time code execution and tool integration through protocols like the Model Context Protocol (MCP), traditional SAST, DAST, and legacy AI safety methods fail to detect modern agentic-AI threats. This paper introduces the LLM Scalability Risk Index (LSRI), a parametric framework and cybersecurity standard for stress-testing autonomous orchestration pipelines. LSRI measures the operational thresholds where load, compounding hallucinations, data poisoning, and adversarial prompt injections including jailbreaking and indirect prompt injection cause security boundaries to fail. Beyond RLHF and RLAIF, we present a Verifiable Root of Trust architecture using cryptographic attestation, semantic policy enforcement, and continuous runtime verification to secure the AI software supply chain. LSRI defends against malicious LoRA adapters, weight tampering, dependency typosquatting, and unsafe model artifacts from public registries such as Hugging Face and GitHub. By replacing post-hoc alignment with verifiable runtime controls, LSRI provides scalable API defense, safer agentic orchestration under heavy cloud workloads, stronger polymorphic malware detection, automated red-teaming, and improved system explainability. Aligned with NIST AI RMF, OWASP Top 10 for LLMs, and ISO 42001, LSRI establishes a deployable compliance baseline for securing generative AI ecosystems including ChatGPT, GPT-4o, Claude 3.5 Sonnet, Copilot, LLaMA, Gemini, and Bedrock. LSRI also supports capital market risk analysis by reducing exposure across technology portfolios linked to indices such as NASDAQ-100, QQQ, and semiconductor-driven supply chains.

Standard evaluations of backdoor attacks on text-to-image (T2I) models primarily measure trigger activation and visual fidelity. We challenge this paradigm, demonstrating that encoder-side poisoning induces persistent, trigger-free semantic corruption that fundamentally reshapes the representation manifold. We trace this vulnerability to a geometric mechanism: a Jacobian-based analysis reveals that backdoors act as low-rank, target-centered deformations that amplify local sensitivity, causing distortion to propagate coherently across semantic neighborhoods. To rigorously quantify this structural degradation, we introduce SEMAD (Semantic Alignment and Drift), a diagnostic framework that measures both internal embedding drift and downstream functional misalignment. Our findings, validated across diffusion and contrastive paradigms, expose the deep structural risks of encoder poisoning and highlight the necessity of geometric audits beyond simple attack success rates.

Membership inference attacks (MIAs) threaten the privacy of machine learning models by revealing whether a data point was used during training. Existing MIAs often assume access to public datasets, shadow models, confidence scores or the training distribution, which makes them vulnerable to defenses like confidence masking. Label-only MIAs avoid these assumptions but require thousands of queries per sample. We propose a cost-effective label-only MIA framework based on transferability and model extraction. Querying the target $M$ with active sampling, perturbation-based selection and synthetic data, we extract a surrogate $S$ on which membership inference is performed offline. This shifts query overhead to a one-time extraction phase. It also removes the restriction that defines the label-only setting: the attacker controls $S$ and can read its posteriors and training trajectory, so attacks that cannot be run against $M$ can be run against $S$. On Location, Purchase and Texas, the strongest attack on $S$ improves AUC over the direct attack on $M$ by $0.9$, $5.6$ and $5.0$ percentage points, and improves the true positive rate at $1\%$ false positive rate by $2.8\times$ to $6.3\times$. We characterize how leakage transfer depends on surrogate fidelity, evaluate standard defenses, and report preliminary results on image datasets.

Privacy preserving machine learning deployments in sensitive deep learning applications; from medical imaging to autonomous systems; increasingly require combining multiple techniques. Yet, practitioners lack systematic guidance to assess the synergistic and non-additive interactions of these hybrid configurations, relying instead on isolated technique analysis that misses critical system level interactions. We introduce PrivacyBench, a benchmarking framework that reveals striking failures in privacy technique combinations with severe deployment implications. Through systematic evaluation across ResNet18 and ViT models on medical datasets, we uncover that FL + DP combinations exhibit severe convergence failure, with accuracy dropping from 98% to 13% while compute costs and energy consumption substantially increase. In contrast, FL + SMPC maintains near-baseline performance with modest overhead. Our framework provides the first systematic platform for evaluating privacy-utility-cost trade-offs through automated YAML configuration, resource monitoring, and reproducible experimental protocols. PrivacyBench enables practitioners to identify problematic technique interactions before deployment, moving privacy-preserving computer vision from ad-hoc evaluation toward principled systems design. These findings demonstrate that privacy techniques cannot be composed arbitrarily and provide critical guidance for robust deployment in resource-constrained environments.

In blockchain applications, transaction confirmation is often treated as usability friction to be minimized or removed. However, confirmation also marks the boundary between deliberation and irreversible commitment, suggesting it may play a functional role in human decision-making. To investigate this tension, we conducted an experiment using a blockchain-based Connect Four game with two interaction modes differing only in authorization flow: manual wallet confirmation (Confirmation Mode) versus auto-authorized delegation (Frictionless Mode). Although participants preferred Frictionless Mode and perceived better performance (N=109), objective performance was worse without confirmation in a counterbalanced deployment (Wave 2: win rate -11.8%, p=0.044; move quality -0.051, p=0.022). Analysis of canceled submissions suggests confirmation can enable pre-submission self-correction (N=66, p=0.005). These findings suggest that transaction confirmation can function as a cognitively meaningful checkpoint rather than mere usability friction, highlighting a trade-off between interaction smoothness and decision quality in irreversible blockchain interactions.

Defending LLMs against adversarial jailbreak attacks remains an open challenge. Existing defenses rely on binary classifiers that fail when adversarial input falls outside the learned decision boundary, and repeated fine-tuning is computationally expensive while potentially degrading model capabilities. We propose MANATEE, an inference-time defense that uses density estimation over a benign representation manifold. MANATEE learns the score function of benign hidden states and uses diffusion to project anomalous representations toward safe regions–requiring no harmful training data and no architectural modifications. Experiments across Mistral-7B-Instruct, Llama-3.1-8B-Instruct, and Gemma-2-9B-it demonstrate that MANATEE reduce Attack Success Rate by up to 100% on certain datasets, while preserving model utility on benign inputs.

Private convolutional neural network (CNN) inference based on secure two-party computation (2PC) suffers from high communication and latency overhead, especially from convolution layers. In this paper, we propose UFO, a quantized 2PC inference framework that jointly optimizes the 2PC protocols and quantization algorithm. UFO features a novel 2PC protocol that systematically combines the efficient Winograd convolution algorithm with quantization to improve inference efficiency. However, we observe that naively combining quantization and Winograd convolution faces the following challenges: 1) From the inference perspective, Winograd transformations introduce extensive additions and require frequent bit width conversions to avoid inference overflow, leading to non-negligible communication overhead; 2) From the training perspective, Winograd transformations introduce weight outliers that make quantization-aware training (QAT) difficult, resulting in inferior model accuracy. To address these challenges, we co-optimize both protocol and algorithm. 1) At the protocol level, we propose a series of graph-level optimizations for 2PC inference to minimize the communication. 2) At the algorithm level, we develop a mixed-precision QAT algorithm based on layer sensitivity to optimize model accuracy given communication constraints. To accommodate the outliers, we further introduce a 2PC-friendly bit re-weighting algorithm to increase the representation range without explicitly increasing bit widths. With extensive experiments, UFO demonstrates 11.7x, 3.6x, and 6.3x communication reduction with 1.29%, 1.16%, and 1.29% higher accuracy compared to state-of-the-art frameworks SiRNN, COINN, and CoPriv, respectively.

Post-Quantum Cryptography (PQC) creates payloads that strain the timing and energy budgets of Personal Area Networks. In post-quantum key exchange (PQKE), this causes severe fragmentation, prolonged radio activity, and high transmission overhead on low-power devices. Prior work optimizes cryptographic computation but largely ignores communication cost. This paper separates computation and communication costs using Bluetooth Low Energy as a representative platform and validates them on real hardware. Results show communication often dominates PQKE energy, exceeding cryptographic cost. Efficient quantum-resilient pairing therefore requires coordinated protocol configuration and lower-layer optimization. This work provides developers a practical way to reason about PQC energy trade-offs and informs the evolution of PAN standards toward quantum-safe operation.

LLM agents rely heavily on high-quality trajectory data to guide their problem-solving behaviors, yet producing such data requires substantial task design, high-capacity model generation, and manual filtering. Despite the high cost of creating these datasets, existing literature has overlooked copyright protection for LLM agent trajectories. This gap leaves creators vulnerable to data theft and makes it difficult to trace misuse or enforce ownership rights. This paper introduces ActHook, the first watermarking method tailored for agent trajectory datasets. Inspired by hook mechanisms in software engineering, ActHook embeds hook actions that are activated by a secret input key and do not alter the original task outcome. Like software execution, LLM agents operate sequentially, allowing hook actions to be inserted at decision points without disrupting task flow. When the activation key is present, an LLM agent trained on watermarked trajectories can produce these hook actions at a significantly higher rate, enabling reliable black-box detection. Experiments on mathematical reasoning, web searching, and software engineering agents show that ActHook achieves an average detection AUC of 94.3 on Qwen-2.5-Coder-7B while incurring negligible performance degradation.

Fuzzing is a powerful technique for finding bugs in software libraries, but scaling it remains difficult. Automated harness generation commits to fixed API sequences at synthesis time, limiting the behaviors each harness can test. Approaches that instead explore new sequences dynamically lack the expressiveness to model real-world usage constraints leading to false positives from straightforward API misuse. We propose stitching, a technique that encodes API usage constraints in pieces that a fuzzer dynamically assembles at runtime. A static type system governs how objects flow between blocks, while a dynamically-checked extrinsic typestate tracks arbitrary metadata across blocks, enabling specifications to express rich semantic constraints such as object state dependencies and cross-function preconditions. This allows a single specification to describe an open-ended space of valid API interactions that the fuzzer explores guided by coverage feedback. We implement stitching in STITCH, using LLMs to automatically configure projects for fuzzing, synthesize a specification, triage crashes, and repair the specification itself. We evaluated STITCH against four state-of-the-art tools on 33 benchmarks, where it achieved the highest code coverage on 21 and found 30 true-positive bugs compared to 10 by all other tools combined, with substantially higher precision (70% vs. 12% for the next-best LLM-based tool). Deployed automatically on 1365 widely used open-source projects, STITCH discovered 131 new bugs across 102 projects, 73 of which have already been patched.