Science & Mathematics

Iris recognition systems are vulnerable to presentation attacks (PAs), including textured contact lenses, print replays, and synthetic irises. Existing PAD methods rely on supervised learning with attack data, limiting generalization to unseen attacks and new sensors. We propose Diff-IrisPAD, an unsupervised framework based on Brownian Bridge Diffusion that models bona fide iris distributions without using attack samples. The method reconstructs clean iris textures from degraded inputs and detects attacks via reconstruction inconsistency using ViT scoring and pixel-level measures. Experiments on a large-scale multi-sensor dataset (125K+ images, NIR and VIS, 8 attack types) show that Diff-IrisPAD achieves an overall ACER of 30.52, significantly outperforming diffusion-based baselines while demonstrating strong generalization to unseen attacks. These results highlight the effectiveness of diffusion-based bona fide modeling for robust and scalable iris PAD.

Perceptive locomotion policies typically observe a dense robot-centric elevation map, although only the terrain beneath and around the next footsteps informs the control decision. Learned encoders compress the whole visual input, leaving the policy to work out which inputs matter while learning the baseline locomotion task. Learning both at once slows training and, without computationally expensive attention mechanisms, makes generalization to new terrain harder. We instead remove redundant inputs from the elevation map before the policy sees it. A lightweight foothold prediction model (FPM) predicts future footholds from proprioception and the velocity command alone, and we sample the elevation map only in small patches around these predicted footholds. We train the prediction model alongside the locomotion policy, periodically retraining it on new policy rollouts as the robot's gait matures. This allows the perception to anticipate where the policy will step, while the policy learns to leverage terrain information around those steps. With the proposed lightweight preprocessing, the policy reaches a given terrain difficulty in roughly a third of the training iterations required with the complete map. It outperforms policies that observe the complete map where footholds are sparse, including in narrow walled corridors, and on unseen terrain, and it matches an attention-based encoder trained at several times the cost. The policy transfers zero-shot to a Unitree G1 humanoid traversing various obstacles, stairs, gaps, and step-downs beyond the training distribution in both indoor and outdoor environments.

Hearing aid compression is typically designed for speech, yet music demands different trade-offs due to its narrower dynamic range (DR) and impulsive transients. This paper investigates whether music fitting benefits from being signal-aware, both in mapping compression demand across signal-audiogram conditions and in tuning its parameters to these combinations rather than inheriting them from speech-derived prescriptions. Compression demand is characterized across eight audiometric profiles, three residual DR severities, and three produced music signals under linear gain. Signal dynamics emerge as a first-order factor: production-compressed music fits the residual DR of the mild-to-moderate profiles evaluated, while dynamically rich recordings challenge even moderate hearing loss. Per-signal optimization of single- and multi-band wide dynamic range compression is then applied to two profiles with the highest compression demand. This optimization improves the Hearing Aid Audio Quality Index over gain-optimized NAL-NL2 baselines while also improving audibility, indicating substantial headroom for music-specific fitting.

Early diagnosis of hepatocellular carcinoma is critical to preserve access to curative treatment options. The Liver Imaging Reporting and Data System (LI-RADS) provides a standardized framework for non-invasive diagnosis on multiphase computed tomography, based on major visual imaging features. However, these features may be difficult to assess consistently, particularly in small liver lesions. We developed dLI-RADS, an interpretable artificial intelligence method designed to estimate the probability of hepatocellular carcinoma from multiphase computed tomography images of manually segmented liver lesions, while remaining aligned with LI-RADS-related imaging concepts. dLI-RADS combines neural network-derived predictions with handcrafted quantitative imaging measurements reflecting LI-RADS major features and integrates these inputs using logistic regression. In this multicenter retrospective study, patients with multiphase computed tomography and at least one liver lesion were included, while lesions larger than 40 mm were excluded. Histology served as the reference standard for hepatocellular carcinoma, and a composite reference standard was used for non-hepatocellular carcinoma lesions. The lesion-level accuracy of dLI-RADS and LI-RADS were compared on the internal evaluation set using 10,000-repetition bootstrap non-inferiority tests, with 5% margins and significance levels. On the external test set, sensitivities were similarly compared. dLI-RADS was non-inferior to LI-RADS for lesion-level accuracy in the internal evaluation set, achieving 80% accuracy compared with 74% for LI-RADS (p = 0.04)...

Edge medical segmentation must reconcile small model size, high accuracy, and fast inference. TernScope addresses this tradeoff through model--hardware co-design, learning a compact segmentation network whose low-bit operators also simplify hardware execution. Split-channel large-kernel blocks preserve spatial context with limited computation, while branch fusion removes redundant inference paths. Ternary Progressive Quantization-aware Training (TPQ) then adapts the fused network from deep to shallow layers, through the decoder, and finally to the input and prediction layers. All 25 convolutions use ternary weights and 8-bit activations, enabling packed storage and multiplier-free computation. The resulting model uses 0.2483 M parameters and 0.0949 GMACs while retaining competitive segmentation accuracy on in-domain and external datasets. Under the same 45 nm technology, TernScope reduces compute-array area by 88.46% compared with the original Gemmini, a systolic-array accelerator with 8-bit $\times$ 8-bit multipliers. These results demonstrate a path from accurate lightweight medical segmentation to compact low-bit hardware.

Object detection models for autonomous driving commonly experience substantial performance drops when deployed under adverse weather due to the domain shift between training data and real-world operating conditions. This degradation is especially evident when models trained on clear-weather images encounter foggy environments with reduced visibility and contrast. To address this challenge, we introduce FogTTA, an online test-time adaptation framework designed to improve the robustness of Transformer-based object detectors in fog. Using RF-DETR as the underlying object detector, FogTTA enables real-time adaptation to the streaming target domain without requiring source data or retraining. The framework follows a teacher-student design, where the deployed model serves as the teacher and generates pseudo labels from weakly augmented target inputs. These predictions are subsequently refined through non-maximum suppression and confidence filtering. The student model then learns from strongly augmented target sample using the Varifocal loss to mitigate pseudo-label noise. The teacher is updated via exponential moving averaging to ensure stable and continuous adaptation. Experiments show that FogTTA outperforms prior baselines, delivering improved detection accuracy and stability while maintaining real-time performance.

Multimodal inputs may be missing or degraded, raising distinct questions of evidence availability and trust. We introduce a reliability-aware Gaussian Product-of-Experts (PoE) in which an input-dependent reliability weight tempers each modality's latent precision, while missing experts are omitted. We characterize how reliability and availability affect posterior information and expert influence. Across two clinical multimodal tasks, the full framework remains competitive with complete inputs and achieves its clearest mean-AUROC advantage under modality absence, particularly on HAIM-MIMIC-MM. Ablations show a substantial gain from missing-aware training and an additional HAIM improvement from reliability weighting under matched PoE, missing-aware, and consistency training. Unseen test-time corruptions are used diagnostically and reveal modality-dependent reliability-weight and latent-uncertainty responses, including failure under severe degradation. Thus, reliability weighting provides an explicit evidence-control mechanism, while learning dependable quality signals remains an open challenge.

An approximate distance oracle (ADO) is a compact data structure that answers distance queries within a multiplicative \emph{stretch} $k$. The classical oracle of Thorup and Zwick~[STOC'01, JACM'05] has stretch $k=2\ell-1$ and uses $O(\ell n^{1+1/\ell})$ space, for every integer $\ell\ge1$, and this trade-off is optimal for dense graphs under the Erdős girth conjecture. However, for sparse graphs ($m=O(n)$), their lower bound yields nothing beyond $Ω(m)$ space. A pivot-based $1$-hop ADO gives every vertex $\tilde{O}(1)$ pivots and returns as an estimate the length of a walk of the form $u\to p_u \to p_v \to v$, where $p_u$ is a pivot of $u$, $p_v$ is a pivot of $v$, and $p_u\to p_v$ is a stored distance. We observe that all the state-of-the-art ADOs with $\tilde{O}(1)$ query time are in fact pivot-based $1$-hop ADOs. Our main result is that, under the Erdős girth conjecture for graphs with constant degree, there are sparse graphs on which every \emph{pivot-based} $1$-hop ADO with stretch $k$ requires $\widetildeΩ(m^{1+2/(k+1)})$ space, for every \emph{real} $k\ge1$. This matches the best known upper bounds with $\tilde{O}(1)$ query time, up to polylogarithmic factors, at every stretch where they are known. We also extend our lower bound to pivot-based ADOs with any constant number $β$ of hops, showing that they require $\widetildeΩ(m^{1+2/(β(k+1))})$ space. In addition, we obtain a tight lower bound for two-hop hopsets. Under the Erdős girth conjecture, for every integer $\ell\ge3$, $(2\ell-3,2)$-hopsets on sparse graphs require size $Ω(m^{1+1/\ell})$, which matches a construction based on the oracles of [Kadria and Roditty, ISAAC'25]. Using the recent explicit high-girth graphs of Double Agent~[doubleAI'26], we also obtain unconditional versions of our ADO lower bounds, with the exponent above $1$ multiplied by $4/5$.

High-performance packet processing frameworks, such as the Data Plane Development Kit (DPDK), are widely used in modern network systems, yet they are rarely integrated into networking curricula. As a result, students often lack exposure to the performance considerations and low-level design choices that shape real-world packet processing systems. We describe a semester-long course project in which students implement a do-it-yourself (DIY) software router using DPDK in a master-level computer networking course. The project guides students through the design and implementation of a multithreaded router, the development of a routing table data structure, and the evaluation of forwarding performance using standard benchmarking tools. We present the project structure and the virtualized infrastructure used to support classes of realistic size. Based on several iterations of the course and survey feedback from participating students, we discuss the observed workload, learning outcomes, and challenges of teaching high-performance packet processing. We release the project material and scripts to recreate the infrastructure, enabling reuse and further refinement by the community.

This paper describes an approach to constructing nowcast and forecast predictive models for the occurrence of high-latitude decameter ionospheric irregularities based on SuperDARN and SECIRA radar data and space weather parameters. The method relies on self-learning pattern recognition in the dynamics of space weather parameters, that influence the irregularities formation. The proposed approach is a data-driven, multiscale method in which the spatial distribution of irregularities is approximated using a neural network of an original architecture, inspired by graph neural networks. The model was trained and tested using data from 11 radars during the period 2010-2025 (SuperDARN: CVW, ICE, ICW, INV, PGR, PYK, RKN, SAS, STO; SECIRA: EKB, MAGW). It is shown that the model achieves optimal performance when using geomagnetic coordinates MLT,MLAT. The optimal set of parameters for the best nowcast and forecast over studied set of parameters is a combination of geomagnetic and solar disturbance indices ($a_{p}$, $K_{p}$, $AE$, F10.7) from ground-based observations, and satellite-measured solar wind and interplanetary magnetic field parameters ($P_{sw}$, $E_{sw}$, $\widetilde{B_{z}}$, $T_{sw}$). Hourly mean values for all the parameters were used in the analysis. The results indicate that acceptable forecast quality is achieved with a lead time of up to 7-9 hours, with the highest quality reached for nowcast (zero lead time). It is shown that the parameters $\widetilde{B_{z}}$ and F10.7 are significant for nowcasting only and can be neglected for other lead times. It is shown that $AE$ and $T_{sw}$ are significant for mid-term forecast (\textgreater 2hours ahead) and can be neglected for short-term predictions. Statistical relationships between the maximum latitude of the equatorial boundary of the irregularities and space weather parameters are also discussed.

Radio Frequency (RF) convergence, the joint use of radio signals for both communication and sensing, is gaining increasing attention in research and industry, mostly for its potential in Internet of Things (IoT) applications. By using RF signals of opportunity, such as those generate by the wireless networks, RF sensing exploits the effects of human bodies on the electromagnetic waves to detect both people presence and their movements without requiring wearable devices. In this paper, we address the problem of estimating the number of people in a monitored area covered by several RF nodes of a wireless network: to this aim, we propose a composite diffraction-based model, namely C-MAM, to capture the cumulative RF attenuation induced by multiple human bodies. Building on this model, we derive the analytical resolvability bounds that quantify the maximum number of distinguishable individuals based on the layout geometry and other physical characteristics of the wireless network. These bounds allow for the pre-deployment assessment of RF sensing performance under different scenarios.

We present an AI-assisted mathematical addendum to Strategic Trip Planning: Striking a Balance Between Competition and Cooperation (Amar, 2018). We reconstruct selected relationships between strategic choice, cooperative stability, participant approval and enforcement. The thesis's Nash-to-core proposition is under-specified until outsider opportunities are stated. We verify two-person implications under fixed outsiders and explicit efficiency conditions. Under isolation, the original example has an empty TU core; its funded counterpart uses an explicit external subsidy. For a monotone resource model we characterize the exact financing gap. A finite negotiation procedure implemented in Lean is proved to terminate with participant approval, joint-cost non-worsening and local optimality against permitted approved moves. Supporting formalizations address bargaining geometry and regret-based efficiency. A resource-game counterinstance shows why monotonicity alone provides no uniform price-of-anarchy bound. Reproducible computations compare a documented heuristic reconstruction with the verified procedure on 44 source-based instances and 96 starting cases from four additional synthetic network families. Acceptance-matched heuristic and randomized pairwise-search controls contextualize the optimization results. The mathematical ingredients are established results; the contribution is a traceable reconstruction linking source statements, explicit assumptions, checked proofs, executable code and retained counterinstances. The study illustrates how AI-assisted investigation can turn engineering intuitions from legacy research into precise mathematical claims whose conditions and computational consequences can be inspected and reproduced.

Indoor localization systems are critical for Industry-4.0 applications. Visible Light Communication (VLC)-based systems offer advantages such as immunity to electromagnetic interference, high accuracy, and energy efficiency, but their performance is hindered by environmental variability across deployment areas. This paper presents a transfer learning (TL) framework for VLC-based indoor localization that enhances cross-area robustness while reducing data requirements. Using real-world measurements from four distinct areas of a Bosch manufacturing facility, we show that fine-tuning a deep neural network (DNN) pre-trained on source area data with only 40% of target-area measurements recovers 97% of full-data performance, thereby reducing the amount of target data needed. We further introduce the Transfer Efficiency Score (TES), a composite metric that identifies the most effective source models for transfer without exhaustive evaluation. Guided by TES, the framework selects the best source model, yielding localization errors as low as 40.9 cm with success rates above 86%. Validation on an unseen area confirms its generalization capability, with a 3.28 cm mean error achieved using the TES-selected source model. Overall, the proposed framework provides a scalable and data-efficient pathway for industrial VLC localization, substantially reducing deployment costs and computational burden while maintaining high accuracy in diverse and dynamic environments.

Quantum computing represents a fundamental shift in computation, promising to solve problems currently intractable for classical computers. However, misconceptions persist; quantum computers are not universally faster, nor do they replace classical systems. Current Noisy Intermediate-Scale Quantum (NISQ) devices face significant challenges, including decoherence, high error rates, and limited scalability. Additionally, leveraging quantum systems requires rethinking traditional computing principles, adopting different programming models, and addressing the complexities of integrating quantum and classical workflows, making their practical use in computer science both demanding and non-trivial. This article provides a system-level perspective on quantum computing for computer scientists, software engineers, and IT professionals, examining its current capabilities, key architectural and software challenges, and potential paths toward quantum advantage in real-world applications within hybrid quantum-classical environments.

Semantic communication jointly optimizes inference accuracy, channel use, and computation, yet the underlying communication program is typically specified before training. This paper presents ComRSI (Communications-Oriented Recursive Self-Improvement), a framework for recursively evolving that program through large-language-model-assisted code edits. A fixed agent modifies the encoder, channel adapters, decoder, loss, and training policy; a protected evaluator screens candidates under prescribed budgets and resource constraints. Accepted code and textual feedback inform subsequent proposals, while candidate weights are trained from scratch. In the original ten-generation CIFAR-100 lineage, eight accepted revisions produce an attention-pooled code achieving 74.3% test accuracy over an Additive White Gaussian Noise (AWGN) channel at 32 complex symbols per image, compared with 67.2% at 256 symbols for the initial design. The program also improves fading-channel accuracy and transfers to Tiny-ImageNet after retraining. A repeat study with ten searches per method compares ComRSI against non-recursive agents with and without textual history. Mean final search scores balancing accuracy, transmission rate, and computational cost are 0.7587, 0.7376, and 0.7089, respectively; the mean numbers of accepted revisions per search are 6.0, 5.0, and 2.1. The advantage over the memoryless agent is supported by a two-sided rank test, whereas the difference from the history-equipped agent remains statistically inconclusive at this sample size. Archived proposals show that pooling can yield a large improvement, while retained history supports subsequent refinement. These results support recursive program optimization within the evaluated setting and expose its remaining rate-robustness and compute-placement trade-offs.