arXiv:2605.07496v2 Announce Type: replace
Abstract: Bird's-eye-view (BEV) images have been widely demonstrated to provide valuable prior information for navigation. Given the global information provided by such views, two key challenges remain: how to fully exploit this information and how to reliably use it during execution. In this paper, we propose a navigation system that uses BEV images as global priors and is designed for ground and near-ground robotic platforms. The system employs an image generation model to interpret human intent from natural language, identify the target destination, and generate traversability masks. During execution, we introduce cross-view localization to align the robot's odometry with the BEV map and mitigate long-term drift in conventional odometry. We conduct extensive benchmark experiments to evaluate the proposed method and further validate it on a UAV platform. Using only a conventional local motion planner, the UAV successfully completes a 160-meter outdoor long-range navigation task. This work demonstrates how the world-understanding capabilities of foundation models can be transferred to embodied navigation, enabling robots to benefit from the strong generalization ability of existing image generation models.
Science Journals
arXiv:2606.06048v2 Announce Type: replace
Abstract: Pathological gait datasets remain scarce due to privacy, recruitment, cost, and movement variability. Our work presents a multimodal LLM-guided framework for pathology-aware 3D gait data synthesis from structured textual descriptions. The proposed method generates fixed-length synthetic skeleton-based gait sequences for pathological gait classification tasks. The framework combines motion tokenisation, pathology-aware language conditioning, LLM-based semantic augmentation, and language-to-gait generation. A key contribution is the proposed pathological tokeniser, which is designed to preserve pathology-specific motion characteristics during discrete representation learning. Experiments suggest that the proposed synthetic sequences improve downstream classification for recurrent classifiers when combined with real data. The best result is obtained using a GRU classifier trained with real and synthetic samples, achieving 92.77\% accuracy under a leave-one-subject-out protocol.
arXiv:2606.06181v2 Announce Type: replace
Abstract: An estimated 5.3 billion mobile phones became electronic waste in 2022. Many of these devices can be repurposed and used in different contexts to extend their lifetime and to reduce ecological impacts. An often overlooked aspect of smartphone reuse is cybersecurity: these devices embed hardware-backed security mechanisms that rely on vendor-controlled provisioning and are designed for a fixed device lifecycle. In this paper, we investigate whether security mechanisms and guarantees remain effective when devices are repurposed outside their original ecosystem. We explore security features in a PinePhone, an open-hardware smartphone, and focus on three core security aspects: boot chain integrity, isolation provided by the Trusted Execution Environment, and the protection of hardware-bound secrets. Our experiments simulate realistic repurposing scenarios and highlight the complexity of reconstructing trust anchors. We generalize our observations to infer requirements for secure repurposing and illustrate how vendor locked mechanisms hinder the repurposing of a majority of discarded devices.
arXiv:2412.19946v2 Announce Type: replace-cross
Abstract: Algebraic theories with dependency between sorts form the structural core of Martin-L\"of type theory and similar systems. Their denotational semantics are typically studied using categorical techniques; many different categorical structures have been introduced to model them (contextual categories, categories with families, display map categories, etc.) Comparisons of these models are scattered throughout the literature, and a detailed, big-picture analysis of their relationships has been lacking.
We aim to provide a clear and comprehensive overview of the relationships between as many such models as possible. Specifically, we take *comprehension categories* as a unifying language and show how almost all established notions of model embed as sub-2-categories (usually full) of the 2-category of comprehension categories.
arXiv:2606.06269v2 Announce Type: replace
Abstract: Animals capable of powered flight range in wingspan from a few hundred microns to a few meters. The inertial turbulence to which these animals are exposed features vortices ranging from a few hundred micrometers to hundreds of kilometers in size. Yet, the impact of ambient turbulence on animal flight is virtually uncharted and most studies on animal flight are conducted in still air or under laminar conditions. Here, we propose a novel parameterization that links animal flight with turbulence, through a proxy for the energy injected into the atmosphere, $E_{sp}=b^3 f^2$, with $f$ the animal's flapping frequency and $b$ the wingspan. We model this parameter using a scaling relation in the shape of a power law $E_{sp} \propto k^\alpha$, with $k=1/b$ the wavenumber corresponding to the animal inverse wingspan. Literature provides four theoretical predictions on the exponent $\alpha$: two connected to aerodynamic and energetic aspects of flight, $\alpha_{aero}=-2$ and $\alpha_{power}=-5/3$, and two linked to physiological limits. Drawing from experimental data of over 400 species spanning 13 insect orders and two vertebrate classes, we recover $\alpha_{power}=-5/3$ as the best scaling relation across the animal kingdom. Grouping per animal clade however reveals a secondary power law with $\alpha=-5/2$ exponent for invertebrate orders, with a family-dependent coefficient. This new scaling relation suggests a yet-unknown universal physical mechanism in insect flight, likely depending on wing morphology and mechanical properties.
arXiv:2606.06506v1 Announce Type: new
Abstract: Grounding systems operating under transient electromagnetic conditions exhibit highly nonlinear behavior influenced by electromagnetic propagation, soil conductivity variations, thermal diffusion, moisture transport, and ionization phenomena. Conventional grounding analyses generally rely on static resistance formulations that neglect the coupled dynamics governing advanced grounding technologies. In particular, rigorous dynamic models describing magneto-active grounding electrode (MAGE) systems remain practically nonexistent within the current scientific literature. A comprehensive nonlinear dynamic model for a MAGE system is therefore developed considering electromagnetic coupling, dynamic grounding impedance behavior, conductivity adaptation, electrothermal effects, moisture-dependent soil dynamics, and nonlinear ionization mechanisms. The proposed formulation was implemented in the MATLAB/Simulink environment and evaluated under combined step and impulsive transient excitation conditions. The obtained results demonstrate that the equivalent grounding impedance evolves dynamically according to environmental and electromagnetic operating conditions, confirming that grounding systems cannot be accurately represented through constant-parameter formulations under transient conditions. Furthermore, the simulations reveal stable nonlinear coupling among electromagnetic, conductive, thermal, and moisture-dependent phenomena. The proposed framework constitutes a proof-of-concept analytical basis for future transient analysis and optimization of advanced magneto-active grounding technologies.
arXiv:2606.06754v1 Announce Type: new
Abstract: We present MADRAG, a training-free framework for analytic essay scoring that combines multi-agent reasoning with retrieval-augmented grounding. Unlike standard LLM-as-judge approaches, which are prone to bias and unstable scoring, MADRAG decomposes evaluation into an interactive process: an Advocate identifies strengths, a Skeptic critiques weaknesses, and a Judge aggregates their arguments into a final score. Crucially, the Judge is augmented with rubric-aligned exemplar retrieval, enabling calibration through comparison with scored examples. Our results show that MADRAG significantly outperforms prompt-based baselines while approaching the performance of supervised systems without requiring task-specific training. Ablation studies demonstrate that retrieval drives calibration gains, while debate improves reasoning on higher-level traits. Our findings highlight the complementary roles of structured interaction and external memory in reliable LLM-based evaluation.
arXiv:2606.06827v1 Announce Type: new
Abstract: Transfer in coordinate networks is often measured by warm-start gain, but whether that gain reflects source-specific structure or generic weight reuse is less clear. We study this question across three implicit neural representation (INR) families, SIREN, ReLU MLPs, and Fourier-feature MLPs, using controlled analytic tests, a 2D lid-driven-cavity Navier--Stokes benchmark, and 1D PDE reference-solution suites for heat, viscous Burgers, and focusing cubic NLS. The analytic tests use independent-seed random controls, while the PDE benchmarks use alternate same-family source controls and auxiliary ablations.
Across settings, transfer magnitude and transfer specificity separate clearly. In a 10-seed controlled 1D geometric test, Fourier Features show the largest structured transfer ($33.1\times$), followed by SIREN ($23.0\times$) and ReLU ($10.7\times$), but ReLU is far more selective: random-control transfer is $0.41\times$ for ReLU versus $14.24\times$ for SIREN. On a controlled two-parameter 1D family, the ranking changes: ReLU gives the clearest structured-versus-control separation at default settings, whereas Fourier Features improve only after bandwidth retuning. In Navier--Stokes and the broader 1D PDE suite, no single architecture dominates every equation, yet the same pattern remains: SIREN often reuses weights broadly, whereas ReLU and, in some equations, Fourier Features are more source-selective. Static diagnostics remain weak, and the heuristic scaling law $A_{\text{transfer}} \propto 1/\Delta t^2$ is rejected in the implemented 1D audit.
These results position transfer specificity as a useful diagnostic for coordinate networks and suggest that architecture selection in scientific machine learning should be evaluated under explicit control conditions, not by transfer magnitude alone.
arXiv:2606.06828v1 Announce Type: new
Abstract: Group Relative Policy Optimization (GRPO) has demonstrated remarkable success in aligning text-to-image (T2I) flow models with human preferences. However, we have identified that the learning loop of current flow-based GRPO is fundamentally decoupled from the learner's current capability, suffering from critical blind spots at both prompt selection and advantage estimation: (i) Existing methods sample prompts randomly, overlooking the substantial impact of data selection on reinforcement learning (RL) efficacy--a factor proven crucial in GRPO for large language models; (ii) They evaluate sample quality solely relying on intra-group statistics, lacking a global perspective to accurately measure true policy improvement. To address these issues, we propose Adaptive GRPO (AdaGRPO), a novel capability-aware RL algorithm tailored for flow models. Specifically, AdaGRPO consists of two principal components: (i) Online Curriculum Filtering Strategy: Dynamically tracks the model's proficiency and adaptively selects prompts that best match its current learning boundary; (ii) Cross-Level Advantage Fusion: Synergistically integrates fine-grained intra-group advantages with macro-level global advantages, providing a comprehensive and unbiased policy evaluation. As a lightweight, plug-and-play module, AdaGRPO can be seamlessly integrated with existing frameworks such as Flow-GRPO, DanceGRPO, and Flow-CPS. Extensive experiments demonstrate that AdaGRPO consistently drives performance gains while significantly stabilizes GRPO training for flow models.
arXiv:2606.06883v1 Announce Type: new
Abstract: Dynamic aperture evaluation relies on long-term tracking, while existing machine-learning surrogates remain difficult to generalize across machines. We demonstrate that coarse-grained dynamic aperture can be learned directly from suitably encoded one-turn maps. By reformulating dynamic-aperture prediction as an image segmentation problem, a deep surrogate model captures the long-term stability topology and transfers to realistic multidimensional Electron-Ion Collider Electron Storage Ring tracking. Failure analysis identifies a challenging resonant regime in which invariant tori are strongly deformed yet remain unbroken. These results establish a proof-of-principle that practical surrogate models can be constructed from one-turn transport information.
arXiv:2606.06982v1 Announce Type: new
Abstract: We present a differentiable framework for end-to-end mutual information (MI) optimization over linear Gaussian directed acyclic graphs (DAGs). The framework targets network-wide design under global constraints, such as a total transmit power budget, and covers MIMO precoding, amplify-and-forward relays, RIS-aided channels, and branching/merging topologies within a common linear Gaussian model. Its core ingredient is a \emph{K-recursion} that analytically propagates all node-pair covariances along the DAG in topological order, including non-adjacent cross-covariances that are necessary for correctly handling branching and merging paths. The resulting covariances yield a closed-form log-determinant expression for the end-to-end MI as a smooth function of the controllable factors. Complex-valued reverse-mode automatic differentiation on this K-recursion then returns the exact Wirtinger gradient at every controllable factor in a single backward sweep, and projected gradient ascent (PGA) is used to maximize the MI under the global constraints. Because no closed-form gradient expression per topology is required, the same topology-agnostic implementation applies to any linear Gaussian DAG. A single topology-agnostic implementation is applied to four representative DAG classes: single-link MIMO, a diamond DAG, a two-hop AF relay, and input-covariance shaping. The same implementation reaches the classical water-filling optimum in the settings where it is available and yields MI improvements in non-single-link topologies without using topology-specific gradient formulas. A further experiment on a multi-layer Gaussian network (11 nodes, 5 layers) illustrates applicability to nontrivial multi-layer topologies for which no closed-form gradient is available.
arXiv:2606.06998v1 Announce Type: new
Abstract: Scattering exceptional points (EPs) are non-Hermitian degeneracies where the eigenvalues and eigenvectors of scattering matrices coalesce, enabling many intriguing phenomena in optical systems. Higher-order scattering EPs are particularly notable for their ultrasensitive response to perturbations, yet achieving flexible, arbitrary-order control remains challenging. Here, we propose a configurable non-Hermitian zero-index material (ZIM) network that enables arbitrary-order scattering EPs, as rigorously proved theoretically and validated numerically. Specifically, we show that in an N-port non-Hermitian ZIM network embedded with loss/gain dopants, the maximum achievable EP order is N, and the order can be flexibly tuned from 2 to N or completely eliminated by adjusting the dopants. Furthermore, we compare conventional coherent perfect absorption with absorbing EPs of different orders. Although both achieve perfect absorption of all incident waves, a second-order EP already outperforms coherent perfect absorption, and higher-order EPs provide further power-law enhancement. These findings establish a pathway toward realizing arbitrary-order EPs in open scattering systems, holding significant promise for advanced sensing applications.
arXiv:2606.07000v1 Announce Type: new
Abstract: Recent post-training methods, particularly Reinforcement Learning with Verifiable Rewards (RLVR), have significantly enhanced the reasoning ability of Large Vision-Language Models (LVLMs). However, the sparse nature of verifiable rewards provides little token-level supervision for failed rollouts, often leading to inefficient exploration in complex multimodal reasoning tasks. Although policy distillation can offer dense guidance, external teacher based methods introduce substantial computational overhead, while answer conditioned tuning methods may expose answer-level information and induce shortcut-like generation behavior. To address these limitations, we propose PTD-PO, a Privileged Tutoring Distillation Policy Optimization framework for RLVR that provides dense guidance without exposing the answer to the student policy. Specifically, PTD-PO constructs structured privileged hints from spatial attention guidance and intermediate textual reasoning steps, and uses them through in-context learning to produce step-wise token-distribution supervision. The student is still optimized under the original answer-free context, and its failed rollouts are aligned with the hint-augmented reference model at the token-distribution level. To further stabilize distillation under the distribution shift between guided and unguided contexts, we introduce a Top-K Jensen-Shannon divergence objective that focuses alignment on informative token probabilities while reducing memory overhead. Experiments on LVLMs ranging from 2B to 8B parameters show that PTD-PO consistently outperforms RLVR and distillation baselines, mitigates entropy collapse, and improves complex multimodal reasoning performance.
arXiv:2606.07119v1 Announce Type: new
Abstract: The current phase of enterprise AI deployment faces a structural failure: organisations are acquiring agentic capability without the infrastructure to govern it. The result is expected to reproduce the error of the first wave of AI deployment: decentralised intelligence without a federation layer leading to a 95% project failure rate. This paper formalises the Three-Ring Architecture as the governing infrastructure of the on-platform organisation. Ring 1 is the existing production architecture; Ring 2 is the M2 federation layer built on strategies-based agentic AI; Ring 3 is the LLM-based frontier intelligence layer. Ring 2 constitutes, in the technically exact sense, the operating system of the agentic enterprise - performing at the organisational level what a computing OS performs at the device level: resource abstraction, process coordination, permission enforcement, and a stable platform for compounding intelligence. A central contribution is the formal distinction between Ring 2 and Ring 3 risk profiles. Strategies-based agents operate within a deterministic framework: their consequences are traceable, their permissions enforceable, their deviations recoverable. LLM-based agents introduce a categorically distinct risk: a non-deterministic actor whose deviations propagate through complex organisational systems without retrospective traceability. Ring 2 is not a useful addition - it is a necessary condition of control and compliance. A further consequence: every improvement in LLM capability is a structural tailwind for this architecture. More capable non-deterministic actors produce larger consequences when they deviate. The governance requirement scales with capability. The architecture has been validated across a decade of deployment in financial services, government, procurement, and compliance among other sectors.
arXiv:2606.07122v1 Announce Type: new
Abstract: We develop a unified DeepONet framework for logarithmically stable infinite-dimensional inverse problems, with inverse acoustic scattering as a model application. The framework is formulated at the operator level by separating the learned inverse map into measurement encoding, finite-dimensional neural approximation, and functional reconstruction components. For inverse maps satisfying a logarithmic stability estimate, we establish quantitative a priori error bounds giving separate estimates for the encoder error, the neural approximation error, and the reconstruction error, thereby characterizing the dependence on the encoder dimension, the network size, and the reconstruction dimension. For comparison, we also record the corresponding Lipschitz-stable estimate arising from the same error decomposition. The abstract theory is then specialized to the recovery of a medium contrast from fixed-frequency far-field measurements. Numerical experiments in two and three dimensions illustrate stable reconstructions under measurement noise.
arXiv:2606.07146v1 Announce Type: new
Abstract: Physics-informed machine learning is often assessed by curve error, although engineering use depends on downstream decisions: ranking candidates, avoiding infeasible designs and limiting regret. We introduce pinn-gym, an open benchmark for material-conditioned lattice design that couples a transparent reduced-order crush-and-impact oracle with five printable polymer cards, dimensionless force-response targets and a protocol spanning curve fidelity, physical admissibility, top-k retrieval and mass regret.
Across per-material, pooled and cross-material settings, low nRMSE is frequently insufficient to identify useful design selections. Physics-informed losses alter trade-offs rather than monotonically improving all metrics, and dimensionless conditioning improves comparability without making transfer symmetric. The benchmark is not a certified material model; within the released oracle, candidate generator and material cards, pinn-gym provides a reproducible testbed for evaluating PIML surrogates as decision systems rather than curve predictors alone.
arXiv:2606.07148v1 Announce Type: new
Abstract: In bipartite graphs, $(\alpha,\beta)$-core is a widely used model for cohesive subgraph mining. Specifically, an $(\alpha,\beta)$-core is a maximal subgraph in which each vertex in the upper layer has degree at least $\alpha$, and each vertex in the lower layer has degree at least $\beta$. The state-of-the-art CPU-based solutions incur extensive costs to construct an index structure for all $\alpha$ and $\beta$ combinations, leading to scalability challenges on large bipartite graphs. Moreover, on-the-fly queries, which aim to determine whether an edge update belongs to a target $(\alpha,\beta)$-core, are essential for real-time applications such as fraud monitoring and recommendation systems. However, existing index-based methods struggle to support such queries at scale due to their high maintenance overhead. In this paper, we investigate how to leverage GPU architectures to enable efficient $(\alpha,\beta)$-core computation and support on-the-fly queries. While GPUs are widely used to accelerate graph processing, their limited memory capacity makes it impractical to store large index structures. To address this issue, we propose GCC, an index-free GPU-based peeling algorithm that accelerates $(\alpha,\beta)$-core computation via warp-centric processing. To further improve efficiency, we develop GCC+, which leverages the nested property of $(\alpha,\beta)$-core with a core-based early pruning strategy. For handling on-the-fly queries, we propose GFQ, a connectivity-aware algorithm that significantly narrows the computation scope by leveraging connected component information, thereby avoiding full-graph peeling. Extensive experiments on 11 datasets demonstrate that our proposed techniques outperform existing CPU-based solutions in terms of both space and time efficiency.
arXiv:2606.07164v1 Announce Type: new
Abstract: We discuss the numerical approximation of the stochastic Cahn--Hilliard equation with a singular double-obstacle potential and multiplicative conservative noise. We propose a regularised fully discrete finite element approximation scheme for the problem and show that it satisfies stability estimates which are uniform with respect to the discretization parameters. We show convergence of the approximation for vanishing discretization parameters towards a regularised version of the singular stochastic Cahn--Hilliard equation by monotonicity arguments. Hence, thanks to a uniform $H^1$-estimate for the regularised problem we show that the regularised solution converges towards the pathwise unique probabilistically strong solution of the original singular stochastic Cahn--Hilliard equation. We conclude by presenting numerical simulations where we compare the regularised numerical approximation to its unregularised counterpart and illustrate the effect of the conservative noise.
arXiv:2606.07237v1 Announce Type: new
Abstract: Large Language Models (LLMs) are increasingly used in healthcare for tasks such as clinical question answering, diagnosis support, and report summarization. Despite their promise, these models remain highly sensitive to subtle prompt perturbations, both lexical and syntactic, posing serious risks in safety-critical clinical applications. In this study, we conduct a systematic sensitivity analysis to evaluate the robustness of both general-purpose (e.g., GPT-3.5, Llama3) and medical-specific LLMs (e.g., ClinicalBERT, BioLlama3, BioBERT) using the MedMCQA benchmark. We categorize perturbations into natural and adversarial types and examine their effect on model consistency, accuracy, and reliability in clinical reasoning tasks. Our findings reveal that medical LLMs are not intrinsically safe. Even minor variations in phrasing can alter clinical advice, and targeted adversarial prompts can provoke harmful outputs. In high-stakes settings like healthcare, such unpredictability is unacceptable-models that change diagnoses due to reworded inputs or hallucinate medications when slightly rephrased cannot be reliably trusted by clinicians. While models tend to show resilience to simple lexical substitutions or paraphrasing, they often break down under syntactic reordering or misleading contextual cues. This fragility is evident across both general-purpose and domain-specific LLMs. Notably, adversarial manipulations can lead to clinically dangerous outputs, such as recommending incorrect dosages or omitting critical findings.
arXiv:2606.07239v1 Announce Type: new
Abstract: The success of generative molecular design hinges on a model's steerability toward high-reward samples. Because many molecular properties are intrinsically linked to molecular size, accurately capturing the joint distribution of properties and the number of atoms is essential. However, current diffusion and flow-based models fix the number of atoms, which ultimately limits their ability to navigate this complex relationship. To address this, we introduce Morph, a flexible-size generative model for conditional and unconditional 3D molecular design based on geometric graphs. By dynamically adapting size, Morph can seamlessly integrate existing structural priors, like scaffolds, and significantly enhances property steering. We show that Morph matches current fixed-size state-of-the-art models while offering the benefit of unparalleled sampling flexibility. We demonstrate out-of-distribution generation in regimes where previous models fail, paving the way for enhanced generative modeling for molecular design.
arXiv:2606.07337v1 Announce Type: new
Abstract: We present Skeletal-Anchored Dual Harmonics (SADH), a novel 3D shape representation that tightly couples local surface geometry with internal meso-skeletal organization. SADH represents a shape as a collection of compact surface patches rooted on internal anchors optimized directly inside the object volume. Each patch is parameterized using a dual-channel spherical harmonic (SH) formulation, where one channel models local radial geometry while the other defines adaptive patch support through a generalized viewing cone. Unlike isotropic primitives such as medial spheres or Gaussian kernels, SH patches directly encode anisotropic local surface geometry together with adaptive spatial support, enabling compact representation of detailed and directionally varying surface regions. Starting from unorganized point clouds, SADH jointly optimizes surface geometry, anchor locations, patch orientations, and structural connectivity through a staged optimization process that progressively forms a coherent meso-skeletal structure. A geodesic anchor graph further preserves structural relationships between neighboring patches. Experiments on complex 3D shapes demonstrate that SADH achieves accurate surface reconstruction together with compact and coherent skeletal organization across a wide range of geometries.
arXiv:2606.07350v1 Announce Type: new
Abstract: This article proposes an analytical method to address the slow convergence of electronic structure calculations caused by the inability of finite one-particle basis sets to describe the electron-electron cusp. An equivalence is made between a calculation using a finite basis set with the physical Coulomb interaction and a calculation using a complete basis set with a model interaction (specifically, the error-function screened Coulomb potential characterized by a range-separation parameter $\mu$). By leveraging the adiabatic connection formalism, a simple, parameter-free correction formula is derived. It depends only on the on-top pair density and a locally defined range-separation parameter ($\mu$) derived from the basis set itself. This `adiabatic connection based basis set error correction' (ABC) is derived from the asymptotic expansion of the wave function at large $\mu$ for small inter-electronic distances. Therefore it is applicable to both ground and excited states without the restriction imposed by the Hohenberg-Kohn theorem. Numerical tests illustrate that the method achieves chemical accuracy using smaller basis sets than typically required.
arXiv:2606.07437v1 Announce Type: new
Abstract: The ISO 26262 standard defines functional safety for road vehicles through risk assessments based on Severity, Exposure, and Controllability, grounded in a human-driven vehicle paradigm. In the context of autonomous vehicles (AVs), the absence of a human driver necessitates revisiting these principles. This paper decomposes the Controllability placeholder into two auditable evidence dimensions of ISO 26262 by introducing two measurable sub-concepts: Transferability and Predictability. Transferability extends Controllability to capture AV systems' ability to hand off control to dedicated fallback safety mechanisms, while Predictability captures how easily external agents can anticipate AV behavior. Predictability is formally defined from human-robot interaction-inspired principles, and a mathematical framework is provided to quantify it. A designed-versus-achievable gap is introduced to distinguish architectural fallback claims from scene-conditioned achievable fallback capability. The proposed metrics align with ISO 26262 and ISO/PAS 21448 (SOTIF), rendering fallback and interaction claims falsifiable and traceable across ODD slices. These dimensions complement rather than replace existing standards, and the enhancements preserve the structure of ISO 26262 while extending its applicability to driverless automated systems operating at SAE Levels 4 and 5.
arXiv:2606.07470v1 Announce Type: new
Abstract: Deploying deep neural network (DNN) inference on low-end edge devices raises two key challenges: protecting model confidentiality against a potentially compromised edge system and enabling verifiable inference without incurring prohibitive overhead. Existing approaches either house partial models and inference software within trusted execution environments (TEEs), resulting in high cost and an application-dependent trusted computing base (TCB), or execute in untrusted environments, providing little security.
In this work, we present VECODI, a framework for verifiable and confidential DNN inference on constrained edge devices. At its core, VECODI introduces SHANGRI-LA, a new execution abstraction on TrustZone-M TEEs that establishes a third runtime environment with privileges strictly between the Secure and Non-Secure Worlds. VECODI leverages SHANGRI-LA to execute untrusted inference code in the Non-Secure World while using minimal application-agnostic Secure-World support to protect model confidentiality and enable verifiability (with respect to proper execution of inference code and model parameters) of inference results. We realize VECODI on a real-world NUCLEO-L552ZE-Q development board and open-source its prototype. Our results show VECODI's small TCB, memory footprint, and runtime overhead, making it a practical option for secure inference in low-end edge devices.
arXiv:2606.07206v1 Announce Type: cross
Abstract: Continuous-variable quantum key distribution (CV-QKD), which uses quadratures of the electromagnetic field, enables practical quantum communication using standard telecommunication technologies. Unidimensional CV-QKD (UD-CVQKD) simplifies the implementation by restricting modulation to a single quadrature. In this work, we experimentally demonstrate a free-space Gaussian-modulated UD-CVQKD system operating under a high detector electronic-noise regime (1.4 shot-noise units). The system employs polarized coherent states with signal and local oscillator co-propagating in the same spatial mode in orthogonal polarizations, ensuring stable interference. System security is analyzed under both untrusted (UTD) and trusted (TD) detector noise models. While no positive secret key rate is obtained under the UTD model, the TD model enables secure key generation over a finite range of modulation variances, highlighting the critical role of detector trust in high-noise conditions. A maximum secret key rate of 270 kbps is achieved at an optimal modulation variance of 11.57. Furthermore, secure operation requires high-transmittance (low-loss) channels under such noise conditions. This study demonstrates the practical feasibility of free-space UD-CVQKD in realistic high electronic-noise detection constraints and highlights detector electronic noise as a key limiting factor in practical systems.