Papers

Sleep provides a nightly window into health by capturing coordinated activity across the brain, heart, muscles and respiratory system. We introduce SleepFM-2, a sleep foundation model developed and evaluated on 282,511 polysomnography recordings from 26 cohorts, including 235,865 used for pretraining. These data span more than two million hours of multimodal physiology. Compared with SleepFM, SleepFM-2 improves disease prediction and sleep scoring, supports arousal, limb movement and respiratory event detection, and transfers to wearable sensing and subjective sleep phenotypes. A model combining its PSG representation with age, sex and BMI met a prespecified discrimination and significance criterion for 188 subsequently recorded EHR phenotypes in two held-out cohorts, including one health system unseen during pretraining. SleepFM-2 also outperformed a 480-feature baseline derived from the same recordings. Its disease scores revealed a reproducible principal component associated with reduced sigma-band spatial coupling and increased hypnodensity entropy. The frozen encoder performed within the observed range of expert scorers for sleep events and transferred to wakeful EEG, headband and in-ear EEG, wrist PPG and wrist accelerometry. It improved sleep staging across six accelerometry cohorts and achieved disease-prediction performance in UK Biobank similar to models pretrained directly on accelerometry. Finally, SleepFM-2 captured aspects of subjective sleep not recovered by conventional PSG summaries, particularly reports of the recorded night. These results show that multimodal sleep physiology can provide a transferable representation of human health across diseases, clinical tasks, sensors and subjective experience.

Reactive sedimentation of mixtures composed by biological and inert solid matter including liquid substrates is modelled by a system of convection-diffusion-reaction equations. In this process, the small solid particles dispersed in the viscous fluid settle due to gravity while simultaneous chemical reactions take place between solid and liquid components (substrates). Among the applications of reactive sedimentation processes, the principal use is in the simulation and control of secondary settling tanks (SSTs) in water resource recovery facilities (WRRFs). The governing equations include a nonlinear strongly degenerate parabolic equation for the total concentration of solids, and transport equations including reaction terms for the solid components and substrates. To accurately approximate the model equations, a high-order finite volume scheme is developed using polynomial reconstructions. Two reconstruction methods are used, a second-order monotonic upstream-centered scheme (MUSCL), and a third-order central weighted essentially non-oscillatory (CWENO3) method. In addition, for the time discretization, we employ strong stability preserving Runge-Kutta (SSPRK) schemes of second- and third-order. The CWENO3 scheme is complemented with limiters, and in the case of zero diffusion, the designed high-order schemes are shown to preserve an invariant-region property. This property is particularly relevant as it ensures that physically relevant numerical solutions are obtained at each time iteration. In addition, high-order approximations for the diffusion terms are proposed to approximate the complete model equations. Simulations of denitrification in SSTs under a continuously operated regime in WRRFs demonstrate the performance of the model and its discretization. Finally, convergence tests show the order of accuracy of the developed scheme and the invariant-region preservation.

Continuous sign language recognition (CSLR) is a key technology for accessibility, yet its development remains limited by the high cost of annotating continuous video streams. Active learning offers a path toward mitigating this cost, but standard acquisition functions are not designed for weakly aligned sign language videos, where sign executions are interleaved with rest poses, irregular pauses, sign-like motion, and temporally redundant frames. This temporal redundancy can undermine sample selection, as acquisition scores may be influenced by timesteps from regions that are not associated with the decoded gloss sequence, distorting the video's estimated informativeness. In this work, we show that modern CSLR models already contain a mechanism for identifying gloss-level temporal evidence: the CTC decoder. Although typically used only during inference, its alignment peaks indicate where the model localizes each predicted gloss in the feature sequence, providing a source of temporal structure for active learning acquisition functions at zero additional labeling cost. Thus, we introduce RAIDAL (Redundancy-Aware Information Density Active Learning), which repurposes the CTC decoder to restrict representation-based scoring to decoder-aligned gloss regions, rather than exposing the acquisition function to the entire unfiltered video. Across three datasets and two architectures, RAIDAL achieves its strongest data-efficiency gains over competing baselines in large-vocabulary, budget-limited settings, while remaining competitive in the smaller-vocabulary, large-budget setting. The code used in this work is publicly available at github.com/verlab/RAIDAL.

When non-expert users ask LLMs for assistance, their queries can often have misconceptions (e.g., "How do I parse XML with regex?"). In such cases, often referred to as the XY-problem, LLMs must identify the misconception ("regex are fragile") and meaningfully direct the user toward a pragmatic solution that will address the root problem implicit in the request ("use an XML parser"). We introduce XYBench, a benchmark of 8,115 such queries, drawn from technical (StackOverflow/StackExchange) and everyday (WikiHow and a manually-curated subset) domains. We design an evaluation paradigm that assesses model responses along three criteria grounded in cooperative response theory: (a) presence and (b) emphasis on pragmatic solutions, and (c) identification of misconceptions. Our experiments show that even the strongest LLMs predominantly answer the literal request (0.75–0.92) and far less often the intended one (0.33–0.71), while substantially lagging behind humans at identifying misconceptions (at most 63% vs. 79–90%). Further, models overwhelmingly prefer pragmatic responses in a multiple choice setting yet consistently fail to generate them. Oracle ablation experiments show that providing explicit user intent at generation time helps; however a large gap remains, suggesting pragmatic redirection is a fundamentally underdeveloped capability in current LLMs.

Web Application Firewalls (WAFs) mainly rely on signatures to detect known attacks, which can leave gaps against modified or previously unseen payloads. Positive security provides a complementary approach by learning legitimate traffic and blocking inputs that fall outside the learned profile. However, learning directly from live traffic can be unsafe when malicious requests contaminate the training data. This paper presents the Whitelisting Autonomous Policy Producer (WAPP), a framework that combines trust filtering, deterministic rule synthesis, confidence scoring, and validation before enforcement. WAPP is evaluated on three controlled applications using a live Coraza and OWASP Core Rule Set (CRS) stack. Results show that, on the tested DVWA username field, unfiltered learning becomes Degraded at 0.2% poisoned traffic and Broken at 0.5%, while the evaluated free text field can admit malicious inputs even without poisoning. On the frozen poisoning dataset, the ablation configuration with all seven candidate signals improves the measured poisoning resilience from 53% to 90%, compared with 62% for the Kruegel–Vigna baseline. The deterministic synthesizer provides attack blocking comparable to the tested language model without model inference cost. WAPP blocks confirmed CRS bypasses on constrained fields, while free text inputs remain a precision challenge that requires character level operator control.

Multi-task biomedical classification requires models to generalize across disparate modalities and class distributions. We study 11 heterogeneous MedMNIST datasets using the harmonic mean of per-task macro-F1. We evaluate three backbones with task-specific linear heads. We identify a resolution domain shift between the MedMNIST API and evaluation environment. Resolving this inconsistency and optimizing architecture-specific regularization substantially improved performance. Our best configuration, a ConvNeXt-Tiny backbone with label smoothing, achieved a leaderboard harmonic-mean macro-F1 of 0.73294 in the Tensor Reloaded: Multi-Task MedMNIST competition, ranking sixth at the close of the official competition phase. Our implementation is publicly available at: https://github.com/GavrilStefan-Dorian/A-Shared-Backbone-Approach-for-Multi-Task-MedMNIST-Classification

Embodied artificial intelligence has historically very much relied on visual perception, leading to a proliferation of multiple vision-centric world models. However, this reliance fails to capture spatial understanding in its entirety and can even present vulnerabilities in environments with visual occlusions, low-light conditions, or blackouts-scenarios, where acoustic information becomes a critical alternative for spatial awareness and navigation. Despite its potential, research into realistic spatial audio and particularly the development of audio-centric world models remains sparse. In this technical report, we introduce BinauralVAE: a flexible, open-source pipeline (https://github.com/Luizerko/BinauralVAE) that explores multiple models for spatialized audio reconstruction, progressing from fundamental baselines to advanced, mathematically grounded architectures. Our approach evaluates various Variational Autoencoder architectures – including complex-valued variants – to learn robust latent representations of binaural signals. Developed alongside AudioWorldSim, our methodology leverages realistic acoustic data captured as a simulated robot navigates an environment. This pipeline establishes a foundation for state representation in a future audio-based world model, designed to map the direct causal connection between navigational actions and their resulting acoustic consequences, and helping to enable sound as an essential complementary modality for spatial knowledge acquisition.

Agentic AI systems execute complex tasks through long-horizon workflows of planning, tool use, and multi-agent coordination. Task failures in these systems often originate from a single step, such as an injected prompt or a flawed plan, and are then amplified through downstream dependencies as the corrupted step propagates across many subsequent agents and tool calls. Existing defenses either target a specific class of attacks or failures, or inspect individual prompts and steps in isolation. Both leave the global dependency structure of a workflow unexamined, and miss the inconsistencies that only emerge when the execution is viewed as a whole. We argue that anomaly detection for agentic AI must reason at the workflow level, where global execution structure exposes signals that local checks cannot see. We present Skynet, a principled workflow-level anomaly detection framework that turns observed multi-agent execution into directed workflow graphs and scores them against learned benign behavior. Skynet jointly models the semantic execution context and the structural organization of inter-agent delegation, tool invocation, and data-flow dependencies, and trains only on benign workflows. Because training never sees attacks or failures, this design naturally extends to zero-day detection: any execution that violates benign workflow regularities surfaces as off-manifold geometry under a single decision rule. We evaluate Skynet on three public agentic safety and failure benchmarks. It sustains high recall together with a sub-1% false positive rate, with per-workflow and per-step latencies low enough for online monitoring of agentic AI runtimes.

AI systems are used worldwide, but they struggle to serve the needs of culturally diverse populations. Prior work on cultural understanding evaluates AI systems on text-only settings or on visual artifact recognition (e.g. foods, clothing). The ability to reason about visually observable behaviors through local social norms, which we call visual norm understanding, remains unexamined. We introduce NormViz-Bench, a high quality, human-validated benchmark of 3,268 contrastive image pairs (6,536 images) spanning 16 countries. Each pair varies only in the culturally relevant behavior (e.g., objects, attributes, spatial relations, and actions) that alters how each image is interpreted. Each image is labeled as conforming to, violating, or irrelevant to local social norms, and pair-level evaluation requires both images to be correctly classified, thereby preventing reliance on superficial visual shortcuts. Even the strongest VLMs, Gemini 3.0 Flash and Qwen2.5 VL 7B, succeed on only 26.6% and 21.6% of pairs, struggling most with identifying violating and culturally benign visual behaviors. Towards bridging this, we introduce NormViz-Train, a training dataset of 64k images paired with explanations. Though absolute performance remains low (<30%), finetuning on NormViz-Train improves pair accuracy relatively by up to 125% and 36% Qwen3-VL 4B and 8B respectively, showing a path forward to teach models to connect visual perception to cultural significance. Together, NormViz-Bench and NormViz-Train establish visual norm understanding as a challenging and consequential frontier for multimodal AI.

The deployment of Hierarchical Federated Learning (HFL) in resource-constrained Internet of Things (IoT) environments requires careful configuration to balance predictive performance with energy consumption and execution time. This challenge is particularly relevant to smart agriculture, where distributed IoT devices can support automated plant disease classification while operating under limited computational and communication resources. This paper presents a constrained Bayesian Optimization framework for the efficient configuration of HFL deployments. The proposed approach jointly explores the deep learning backbone architecture, aggregation strategy, and number of communication rounds, while the federation size is determined according to the spatial coverage requirements of the agricultural deployment. A weighted objective function captures user-defined trade-offs among energy consumption, execution time, and predictive performance, while explicit constraints ensure compliance with deployment-specific resource and accuracy requirements. The framework is evaluated on an IoT-based plant disease classification task considering multiple deep learning architectures, federated aggregation strategies, and communication-round settings. Experimental results across 30 independent optimization runs show that the proposed approach explores only 11.11% of the search space, while consistently identifying solutions within 1% of the exhaustive-search optimum, with a mean optimality gap of only 0.056%.

This paper examines the theory of Invariant Structural Learning (ISL), which proposes a non-optimization approach to concept formation. Learning is interpreted as convergence to structural attractors in a hypergraph space, rather than as the minimization of a global loss function. The paper presents the ISL model, including its mathematical formalization, computational verification, and a hypothetical neurobiological interpretation. The mathematical section introduces the formal apparatus of the structural reduction process and proves its finite convergence, the existence and uniqueness of class structural attractors, and the self-organization of attractor maps. The computational section demonstrates the feasibility of the proposed approach on classical image recognition tasks, utilizing the proposed learning mechanism without backpropagation and with extremely small training datasets. Finally, the neurobiological section formulates hypotheses regarding the possible implementation of structural attractors in dendritic trees, neural coding as a projection of internal attractor dynamics, and the development of neural architectures supporting the proposed learning concept. These hypotheses are discussed in the context of modern experimental data in the fields of dendritic computations, synaptic plasticity, and the structural organization of neural circuits. The proposed neurobiological mechanisms are presented as testable hypotheses rather than established biological facts. The results demonstrate the mathematical consistency and computational feasibility of the proposed model, while the neurobiological hypotheses outline potential directions for its experimental verification.

The next generation of artificial intelligence systems will likely be natively temporal and multimodal in both inputs and outputs: able to converse, perceive, predict, reason, and synthesize through a shared world representation. Realizing this requires a temporal perceptual substrate integrating sensory streams, language, and structured outputs within a multimodal world model. We seek a generalist open-world perceptual system that represents biological forms, natural physical structures, and artifacts, and their interactions, as a coherent, temporally persistent process. The model should infer geometry, articulation, semantics, interaction structure, and uncertainty from raw multimodal streams; maintain identity through occlusion and viewpoint change; generalize across species, forms, mechanisms, and materials; and abstain or expand its ontology when encountering the unknown. The objective is a structured world state supporting understanding, prediction, counterfactual reasoning, and controllable synthesis. Recent work suggests that some cross-modal and reasoning-like capabilities can emerge from large-scale generative video pretraining, reminiscent of language-model scaling. Yet these capabilities are often accessed through language probes or expressed through photorealistic video, leaving explicit semantic, geometric, or temporal structure largely unexposed. This paper articulates an alternative and complementary paradigm: perception and synthesis as distinct conditionings within a shared Generalist Open-World Temporal Perception Architecture (GOWTPA). Recognition, structured prediction, and simulation arise as different conditionings of the same generative substrate, while reasoning and embodiment-specific policies build upon the resulting world state. This positions generalist temporal perception as a potential foundation layer for broader multimodal intelligence and physical AI.

Active mapping requires a robot to select camera viewpoints that efficiently reconstruct an unknown 3D scene. To reason about unobserved regions, recent systems use pretrained occupancy networks as world models that complete missing geometry. The predicted structure contributes to expected coverage gain and constrains feasible robot motion. Consequently, occupancy errors can change both what the robot chooses to explore and where it is able to move. We diagnose these effects by holding the planner fixed and varying only the occupancy representation provided to it. We consider planning without completion, with learned occupancy, with false positives removed by a ground truth oracle, with false negatives restored by an oracle, and with ground truth occupancy. Our experiments show that correcting false positives or false negatives alone does not consistently improve final coverage. This finding reveals a gap between occupancy accuracy and downstream planning performance. Ground truth occupancy provides a much larger improvement in coverage efficiency than in endpoint coverage, suggesting that planning and reachability remain important bottlenecks even when the geometric world model is accurate. Based on these findings, we introduce a dynamic filtering strategy that preserves predictions in unexplored space while suppressing repeatedly unsupported occupancy using online observations. Preliminary examples show that this strategy can redirect viewpoint selection toward reachable surfaces that would otherwise remain unobserved.

Decision-time search in perfect and imperfect information games with enumerable belief states are effective methods for game AI. Collectible card games are imperfect information games with large belief states. Legends of Code and Magic is a collectible card game competition where the belief states are $2^{101}$. The Legends of Code and Magic (LoCM) champion, ByteRL, plays with no search. Other works claim sound enumeration-based search is unusable in the genre due to the number of belief states. We measured three previously defined properties that predict where theoretically unsound perfect information Monte Carlo's defects are cheap and found LoCM sits in the favorable region. Starting with imitation learning of the runner-up policy, NeteaseOPD, we created a policy and value feed-forward network. Our agent searches over worlds sampled from a prior over the opponent's deck built from the runner-up's drafts. Using our strictest configuration in the battle phase we beat ByteRL with a win percentage of 51.35% 95% CI [50.37, 52.33], over 10,000 pre-registered games using the LoCM official referee and time limit. Search is not a minor factor on the matchup between our agent and ByteRL. Without search this agent scores 26.8% and adding search adds +24.6 points. Unsound search in imperfect information games could be exploitable. We replicate a published best-response attack against ByteRL. We then apply the same attack protocol to two search configurations of our agent, and each one resists it better than ByteRL at every iteration. In LoCM unsound search gives us a stronger and more resilient agent.

An open, networked web will allow agents to run frozen models from multiple vendors, keep their history private, and teach each other which tool to call and when. Flat text (prompts, example pools) makes it difficult for the protocol to distinguish between noise statistics, merging rules, and documentation. Weights and adapters cannot transfer that knowledge between platforms. We suggest sharing typed federated artifacts, schema-validated objects with well-defined fields for per-field privacy (described here, but measured), dispute resolution, and cross-model transfer, and instantiating them as SYNAPSE1, a common tool-routing knowledge. After deleting 192 garbage entries and 1,916 training items that duplicate or almost duplicate test queries, a federated compendium routes within 1.1 points of a centralized one at 20 MB of JSON per client each round on StableToolBench (3,180 tools). The same experience merged and shown to the router as typed fields rather than one flat string is worth 8.5 points on clean data and 7.4 under 60% injected contradiction. Crossing merge and rendering shows the halves are inseparable (the typed merge shown flat is the worst arm), while three conflict policies are indistinguishable, so the conflict log that motivated this work is not the On {\tau}-bench retail, each compendium arm improves GPT-4o agents' per-step tool-call accuracy by at least 6.7 points, attributed to format rather than federated experience. Two cautionary findings conclude the paper: on a topic-labeled math proxy and StableToolBench, a TF-IDF classifier over the same labeled experience beats every LLM routing arm (by 48 and 26 points, mostly retrieval recall) because the benchmark's pool holds labeled queries for every supposedly unseen tool and every test query verbatim before our filter. It cannot measure routing to tools without labels, which routing exists for.

Urban traffic monitoring plays a critical role in safety analysis, congestion management, and incident response. The growing deployment of robotaxis creates a new opportunity for network-level traffic monitoring. Although robotaxis are primarily designed to serve passengers, they can also be leveraged as drive-by sensors to collect traffic data. Compared to conventional infrastructure sensors or probe vehicles, a fleet of robotaxis forms a cooperative perception environment, which can collectively gather spatially and temporally continuous traffic information. This paper proposes a novel dynamic robotaxi routing framework that explicitly incorporates traffic monitoring tasks as an objective. The framework introduces: (1) a cell-based network representation that aligns with sensing capabilities of robotaxis; (2) a cell-level monitoring metric to quantify spatiotemporal robotaxi coverage; and (3) a mixed-integer linear programming (MILP) formulation that jointly minimizes time-dependent travel time and maximizes traffic monitoring performance. A 5 by 5 urban grid network is built in SUMO to evaluate the framework under three robotaxi market penetration rates (2%, 5%, and 10%) with a range of objective weight combinations. Results show that incorporating spatiotemporal network coverage in the objective function can effectively improve the traffic monitoring performance. Interestingly, with appropriate weights between the two objectives, monitoring performance and robotaxi average speed can be improved simultaneously. This suggests better network monitoring leads to more accurate traffic state prediction and improved mobility. This win-win situation could incentivize robotaxi operators to contribute their vehicles as drive-by sensors for traffic monitoring.

Wireless networks face ever expanding throughput demands from heterogeneous, high-density user populations, requiring beamforming algorithms that adapt to channel conditions with low latency. Validating such algorithms requires either costly over-the-air testbeds or simulation environments that lack the timing and resource constraints of real hardware, leaving a gap between algorithm design and hardware-realizable deployment. This work presents a frame-based hardware-in-the-loop (HIL) testbed that closes that gap by coupling an FPGA-based implementation of OFDM Waveforms with MU-MIMO beamforming to NVIDIA Sionna's ray-tracing channel simulator, enabling a physical base-station architecture to transmit and receive against a Sionna-rendered digital-twin propagation environment. Unlike prior work that validates beamforming algorithms either purely in simulation or on full RF testbeds, this architecture allows beamforming logic running on actual FPGA fabric to be evaluated under realistic, controllable, and repeatable channel conditions, including UE mobility and site-specific multi-path, without requiring an anechoic chamber or live RF front end. We detail the FPGA OFDM transmit/receive pipeline, the synchronization and data interface between the FPGA and the Sionna environment, and validation of signal quality under AWGN and ray-traced channel conditions, establishing this testbed as a platform for hardware-validated beamforming research.

GEO (generative engine optimization) visibility scores aggregate source appearances, citations, or brand mentions in generated answers. The prompt corpus selects the situations evaluated, while weights determine their relative importance. Together they define an "answer market" that need not represent actual user demand. Prompt wording can alter retrieval, competing sources, and generated answers. Scoring then requires identifying the appearances, citations, or mentions of interest. If a language model performs this task, its instruction can change the score assigned to an unchanged answer. Our critical survey examines how these choices help define what a GEO score measures. It draws on research into whether indicators measure the intended phenomenon, total survey error, and information retrieval evaluation. The framework specifies situation annotation, prompt formulations, execution conditions, weights, and scoring rules. When weights are unknown or remain to be chosen, the framework reports sets of admissible scores. It distinguishes values compatible with data and assumptions about a target population (partial identification) from variation across weighting conventions (normative sensitivity). A citation alone does not establish a source's contribution. The article defines a comparison of answers generated with and without a source in a controlled documentary context, distinct from an intervention on the full engine with competing sources. The framework is supported by reproducible calculations. No new experiments are reported; its general empirical validity remains to be assessed.

Optical coherence tomography (OCT) offers compact, contactless, micron-scale imaging suitable for intraoperative guidance, but native OCT volumes are discretely sampled, anisotropic, and currently inefficient for downstream geometric reasoning and robot integration. We present OCTN (pronounced "octane"), an implicit neural representation framework that converts volumetric OCT scans into a continuous, differentiable, and spatially faithful tissue-intensity field. OCTN uses a two-stage hybrid training strategy that combines supervision from acquired voxels with inter-slice interpolations, preserving B-scan fidelity while improving continuity in sparsely sampled regions. For versatility, we first show that OCTN enables fast volumetric reasoning by storing the learned tissue representation natively on the GPU, supporting intensity-based spatial queries with up to 43x speedup over conventional CPU processing. We then demonstrate OCTN-enabled OCT-guided robotic laser surgery where the continuous tissue representation supports implicit surface discovery and surface-constrained path planning via multiple optimization strategies, including Newton- and SGD-based optimization. Next, OCTN enables reconstruction of dense volumetric structure from sparsely acquired B-scans, while reducing acquisition time by 4x and preserving clinically relevant structures. Across the newly generated Duke TissueOCT dataset and public OCT datasets, OCTN achieves robust, high-fidelity reconstruction with PSNR > 30 dB and training time < 10 s, while preserving surface consistency within 10 $\mu$m Chamfer distance relative to baseline reconstruction. The TissueOCT dataset and code are publicly available at raprakashvi.github.io/octn

Modern stereo matching models fail when disparity crosses zero, with end-point error (EPE) rising by 4.6-37$\times$. Yet stereoscopic content, from cinema 3D to VR, routinely contains objects behind the zero-disparity plane (ZDP), corresponding to negative disparities. The blind spot cascades through datasets, architectures, and evaluation protocols, all of which inherit the non-negative geometry. Rectified parallel cameras place ZDP at infinity, so every finite depth yields $d=fB/z \ge 0$ by construction, and nothing within the standard pipeline can violate, or even measure, a negative disparity. To measure it, we propose ZDPShift, a benchmark of $21{,}495$ stereo pairs from seven cinematographer-authored open movies, each frame rendered at five zero-disparity-plane positions with dense signed ground truth. Six state-of-the-art image and video stereo matching models collapse once the plane moves. On identical scene content, FoundationStereo goes from $2.24$ px EPE to $75.33$px, with every backbone leaving roughly half of all pixels exceeding a three-pixel disparity error. What is missing, however, is not the underlying matching capability. % The capability itself, however, is already present. Training on supervision synthesized from SceneFlow, which adds no new data or parameters, keeps the error flat across the signed range. Training only the decoder, with the pretrained matching features frozen, performs comparably across all six backbones, with EPE jittering within $0.2$px. Thus, the pretrained features already extend to the negative regime they were never trained on, and only the output convention discarded it. Meanwhile, positive-regime accuracy on KITTI, Middlebury, ETH3D, and Sintel is largely preserved.

We investigate secure communication over a classical multiple-access wiretap channel (MAC-WTC), specifically exploring the benefit of shared entanglement between the transmitters. Under the strict semantic security criterion, we derive an achievable rate region and a regularized expression for the secrecy capacity. We further establish a single-letter upper bound on the secure capacity of the MAC-WTC with entangled transmitters. Our results demonstrate that entanglement strictly enlarges the secrecy capacity of MAC-WTC compared to sharing only classical correlated randomness, a finding we illustrate using a pseudo-telepathy game example. Finally, we establish new strong soft-covering lemmas for the output statistics of multipleaccess channels (MACs) with entangled transmitters. Our results generalize existing results for non-entangled systems.

In this work, we comprehensively evaluate three popular feature-extraction paradigms in AI-based neuroimaging modeling: (1) computation of anatomical surfaces and volumes, (2) supervised learning with convolutional neural networks (CNNs), and (3) unsupervised pretraining of vision transformer (ViT) foundation models, followed by supervised finetuning. Our study is based on 18 publicly available datasets containing 3D structural T1-weighted MRI scans from approximately 80,000 participants across seven distinct clinical tasks. We observe that a linear model based on anatomical features matches the diagnostic performance of complex nonlinear features learned by sophisticated AI frameworks, including foundation models trained on thousands of scans. Conversely, CNNs and pretrained ViTs learn features that implicitly capture relevant anatomical information, bypassing the need for explicit feature extraction. Building upon these insights, we propose Anatomy Segmentation Pretraining (ASP), a novel method to incorporate anatomical information during foundation-model pretraining, which outperforms existing models in biological age estimation.

Training strategy, namely whether to retrain from scratch or fine-tune from the previous checkpoint, is an overlooked decision variable in active learning. We show that this choice has exploitable structure: retraining is most useful in early rounds, when each batch can substantially reshape the labeled distribution, while fine-tuning becomes safer once the model trajectory stabilizes. We propose HybridAL, an adaptive training schedule that monitors an online stabilization signal and switches from retraining to fine-tuning after sustained stabilization. Two complementary signals, spectral exponent change $\Delta\alpha$ (weight-based) and accuracy change $\Delta$Acc (validation-based), span different points on the time-calibration trade-off. Across three encoder backbones and six text-classification tasks (five seeds each), HybridAL keeps endpoint macro-F1 non-inferior to retraining and fine-tuning at a 0.010 margin, saves up to 49% of retraining time, and recovers a substantial fraction of retraining's calibration advantage as measured by negative log-likelihood (NLL). Compared with schedules that switch at a pre-committed round, HybridAL obtains lower NLL at moderate additional cost, showing that trajectory-dependent switching provides a stronger time-calibration trade-off than fixed early switching.

Combat aircraft and other modern military platforms field two co-located RF assets: a radar for sensing and a dedicated radio for communication. In contested battlefield environments, however, the dedicated communication path is highly likely to be targeted and disrupted by jamming, blockage, or physical damage, while the radar remains fully operational. A natural fallback is to repurpose the surviving radar aperture as an emergency communication channel. The difficulty is that radar is purposely engineered for sensing rather than communication, so loading any information onto its waveform perturbs the beampattern and inevitably erodes sensing performance: the more bits we push, the more sensing we lose. We identify the core problem as enabling emergency communication via a radar-centric integrated sensing and communication (ISAC) link while preserving the radar's primary sensing function. Therefore, we propose a semantic radar-centric ISAC framework that pairs a learned vector-quantized semantic encoder, which compresses task-relevant content into a few tens to a few hundred bits, with a wideband phased-MIMO radar restricted to a conservative spatial-spectral configuration subset that respects a bounded sensing-degradation budget. Experiments on SAR target classification and text classification show that our framework supports reliable emergency communication for downstream tasks while keeping the radar beampattern almost unchanged.

Artificial intelligence is being rapidly incorporated into traditional search systems, yet scant work audits the information disparities across platforms, geography, and languages. We address this gap by comparing Google and Baidu's AI Overview systems for health queries, and measure informational anchors that emerge. Auditing 1,920 health queries across 12 countries and 4 languages, we find that Google and Baidu exhibit vertical integration, routing users toward their own company platforms in AI Overviews rather than a diverse set of primary sources. Smaller, lower-localization countries receive fewer domestically sourced references for health queries. Issuing the same query in a country's official language rather than English raises the share of locally sourced citations approximately 3.5- to 13.5-fold. Comparing queries across health topics of varying severity and controversy, including Traditional Chinese Medicine as an example, we also show that health disclaimers are multidimensional and vary across language and culture. We discuss how generative search influences access to health information, and the urgent need for culturally-aware oversight of these systems that influence critical health decisions.

Multi-chiplet photonic neural network accelerators (MCPNAs) combine photonic computation, photonic communica-tion, and heterogeneous chiplet integration to enable scalable and energy-efficient AI acceleration. However, their distributed archi-tecture and reliance on third-party chiplets introduce significant hardware security risks. This paper examines Hardware Trojan (HT) threats to MCPNAs across three dimensions: confidentiality, integrity, and availability.

Android agents using a hybrid action space that combines GUI actions and tool actions (e.g., accessing application data via APIs) remain largely underexplored, mainly due to the excessive effort required to create tools. To address this gap, we introduce DroidTool, a framework for augmenting the agents with self-generated tools, which are realized as Python functions operating on application states (e.g., a database). To create tools with minimal human labor, DroidTool employs an agentic workflow featuring stages: proposal, implementation, test generation and execution, and repair. Notably, when testing the created tools for verification, it constructs relational tests across relevant tools for natural preparation of appropriate test preconditions and improved test coverage, rather than testing each tool separately. The GUI agents augmented with the generated tools achieved approximately 4.47%p higher performance with approximately 20.05% fewer interactions than the GUI-only agents, averaged across representative benchmarks: AndroidWorld, B-MoCA, and MobileSafetyBench.

Register-Transfer Level (RTL) simulation is widely used to test hardware before it is fabricated. To allow testing for security related information flow properties, such as confidentiality and integrity, taint logic can be automatically added to the design to track how information flows through it. However, taint logic instrumented by the state-of-the-art, such as CellIFT, makes simulation-based testing prohibitively expensive: On our evaluation of Mega-BOOM (136K cells), it increases the instrumented design to 5.81x the original cell count and causes a 143.72x simulation slowdown. The taint logic could be simplified to improve simulation speed, but it will inevitably trade off its precision. This lightweight, imprecise taint logic will introduce false positives and may eventually result in even more overhead to check these false positives. This paper explores the research question of where precision is actually needed in the design to overcome the overhead of false positives. It presents CEGAR-T, a framework that automatically synthesizes taint logic that minimizes the taint-logic instrumentation overhead while guaranteeing no false positives (relative to the precise CellIFT baseline). We have implemented CEGAR-T and evaluated it on the safe instruction set problem for timing side-channel security across open-source RISC-V cores. Over all evaluated cores, CEGAR-T reduces both instrumentation and simulation overhead, in geometric-mean, from 5.64x to 1.42x and from 34.65x to 1.79x, respectively, without compromising the precision benefit of the CellIFT baseline.

In secure distributed matrix multiplication, a user wishes to compute the product of two matrices with the assistance of servers, in such a way that any $T$ colluding servers learn nothing about either matrix. Degree tables are a combinatorial tool for constructing polynomial codes for this problem and underlie several state-of-the-art schemes, including $\mathsf{GASP}_r$, $\mathsf{GASP}_{r,s}$, and $\mathsf{DOG}_{r,s}$. We introduce a periodic-gap framework for constructing degree tables that contains these three families as special cases and leads to two new constructions, $\mathsf{SHIFT}_{r,s}$ and $\mathsf{COVER}_r$. We determine their exact recovery thresholds and show that, in many cases, they outperform the current state of the art. We also prove a new lower bound on the recovery threshold of ordinary integer degree tables. In the balanced case, in which the partitioning parameters and the security parameter are equal, we sharpen this bound to match $\mathsf{SHIFT}_{r,s}$ up to lower-order terms, showing that it is asymptotically optimal among degree-table constructions.

Automatic Pronunciation Assessment (APA) systems have achieved strong performance with transformer-based models and self-supervised speech representations. However, most methods rely only on acoustic signals and overlook temporal synchronization between speech and articulatory movements, limiting diagnostic feedback on timing mismatches important for L2 pronunciation training. We propose an interpretable audio-visual synchronization framework that explicitly models speech-lip temporal alignment through feature encoding, cross-attention fusion, lag estimation, stability quantification, and visualization. The framework introduces frame-level lag trajectories and a Lag Stability Index (LSI) to quantify synchronization robustness. We also interviewed 30 participants, including 10 instructors and 20 students with diverse first-language backgrounds, to assess its effectiveness. By transforming implicit alignment into interpretable representations, the framework connects automatic scoring with actionable Computer-Aided Pronunciation Training feedback. Datasets and supplemental materials are available at https://www.robots.ox.ac.uk/~vgg/data/lip_reading/.