Papers

We scale our conventional sub-150M pretraining recipe from 53.5M to 109.7M parameters, holding the method fixed (Qwen3-style decoder with grouped-query attention, RoPE, SwiGLU, RMSNorm, QK-Norm, and a z-loss; FineWeb-Edu data) and changing only the geometry to a deep-and-thin 23-layer x 576-hidden design. The larger model improves across the board – BLiMP 78.1 -> 81.3, ARC-Easy 51.4 -> 52.5, WikiText-2 byte-perplexity 2.04 -> 1.95 – and its 81.3% BLiMP essentially matches GPT-X2-125M (81.28) at about 12% fewer parameters. Notably the 110M model achieves this on fewer training tokens (about 8B vs 12B), so the gain is attributable to capacity and depth, not more data. Both models are deliberately conventional; this report is a clean scaling control and the baseline rung (R0) of an ablation study of what further improves models in this regime. An ablation ladder follows: value residuals (R1) and the Muon optimizer (R2) lift ARC-Easy by a cumulative +3.6 (52.5 -> 56.1) at a near-flat BLiMP and are kept; a diverse data blend (R3) and two logit-distillation settings (R4a/R4b) are not kept – honest negatives. R3 pins ARC-Easy to FineWeb-Edu's educational filtering rather than raw diversity; distillation from a 1.7B teacher can reach the class-leading ARC-Easy (56.99, matching GPT-X2-125M) but only at a perplexity cost that dialing KD down then erases – so R2 remains the best kept stack.

Metastability—a phenomenon where systems remain trapped in quasi-stable states before abruptly transitioning under rare perturbations—is ubiquitous in physical systems. Although metastability is a widely observed phenomenon, its identification and analysis present significant challenges. To address these challenges, we propose a novel framework for analyzing metastability using Koopman theory. We use a finite set of system trajectories to learn a representation of the dynamics that defines a latent space in which the system evolves linearly, thereby enabling a systematic characterization of metastable behavior through the spectral properties of the linear mapping. Empirical evaluations demonstrate that our approach is capable of anticipating metastable behavior significantly earlier than its actual manifestation, even with $10\%$ of the simulation duration. Moreover, we establish that the dominant eigenvalue of the learned Koopman matrix in the latent space serves as a critical indicator for detecting metastability across both single-server and multi-server configurations.

Bayesian quantum tomography requires efficient inference while preserving a posterior fixed by the prior and Born likelihood. Learned transport provides fast amortized samples, but reward tuning can reshape the generated distribution rather than improve exploration of this fixed target. We introduce GRPO-QPS, a target-preserving framework in which GRPO learns proposal behavior and an exact Metropolis correction preserves the posterior after training. Across the evaluated reconstruction benchmarks, GRPO-QPS improves over BuresTomFlow and Flow-GRPO on thermal, cat, Dicke, and cluster families, and it closely matches an exact two-qubit reference posterior. Tuned conventional MCMC is slightly stronger on several original continuous benchmarks where the available fixed proposals already match the posterior geometry well. To test whether this reflects a fundamental limitation of learned exploration, we evaluate a more challenging multimodal thermal posterior. At six qubits and 800 shots, the learned proposal achieves a minimum effective sample size of 102 per $1{,}000$ likelihood calls, compared with 28 for prior independence, 27 for a tuned fixed mixture, and 20 for Haario adaptive Metropolis. A record-conditioned policy also transfers to unseen 3,000-shot records, matching or exceeding the strongest conventional baseline in all nine held-out seed-record comparisons. These results show that GRPO-QPS combines target-preserving Bayesian inference with broad gains over learned transport baselines and a sampling advantage when efficient exploration requires proposal geometry beyond the evaluated conventional kernels.

Interaction paradigms used in robot-assisted autism intervention have historically employed robots as teachers, clinical assistants, or more-abled peers to promote a variety of social skills. These modalities often leverage the expertise of trained practitioners to ensure that child-robot interactions are productive or clinically grounded to yield positive therapeutic benefits for children across the autism spectrum. Yet, despite the fact that the majority of children with autism spectrum disorder (ASD) attend mainstream schools and spend 80% or more of their time in the general classroom [27], there is a paucity of research incorporating validated classroom teaching pedagogies into robot-assisted autism interventions. In this work, we introduce a novel teaching methodology for advancing social skills in school-aged children with ASD. We evaluate the effectiveness of a novel robot-assisted autism intervention which incorporates the learning-by-teaching pedagogy and explores the comparative benefits of employing a robot versus a human confederate for improved performance on a set of social skills tasks. Results show that 80% of study participants performed better in the robot condition (mean performance in the robot condition=63%, mean performance in the confederate condition=37%), irrespective of the scenario order. Further, 90% of all participants were significantly more engaged in the robot condition (mean engagement: robot=61%, confederate=32%) and, while the effect did not result in the confederate condition, analyses indicate that overall engagement in the robot condition contributed to improved performance. These results suggest that robots employed in a learning-by-teaching context may help enhance engagement and improve performance on a simple social skills task for children with ASD.

Immune therapies act across cell-intrinsic programs, tissue ecosystems, and patient-specific immune states, yet most predictors address these scales separately. We used a governed evolutionary AI Scientist to construct the Immune World Model, an action-conditioned model that learns how interventions move immune states across cellular, tissue, and individual levels. The Immune World Model–building Scientist searched candidate architectures and workflows, and the resulting world model was frozen before independent confirmation. The frozen model generalized to unseen interventions and biological contexts, recovered intervention-specific cellular programs, integrated cell and tissue information to improve ecosystem and patient-response prediction, and forecast unseen perturbation combinations. Immune World Model–guided analysis then combined measured perturbations with cross-axis inference to nominate IL-36$\gamma$ plus SIRP$\alpha$ inhibition as a complementary-axis therapeutic hypothesis, whereas a governed self-correction audit rejected every screened cytokine pair. The Immune World Model provides a framework for multiscale immune simulation that connects AI Scientist-driven model construction, intervention forecasting, and the generation of prospectively testable therapeutic hypotheses.

A core goal of efficient reasoning is to improve the accuracy-efficiency frontier. However, jointly improving reasoning accuracy and inference efficiency can be challenging, as the two objectives can favor different reasoning behaviors. Independently post-trained models already offer distinct strengths in accuracy and efficiency. We introduce Lightning Weave, a post-training framework that extracts and composes these independently learned capabilities in a single student through on-policy distillation. Each acquired capability is represented by the policy shift from the model before post-training to the resulting specialist. Lightning Weave combines aligned log-ratio shifts at shared student token states and uses Tilted-Target DOPD to convert the cached signals into a stable learning target. Each anchor pair scores the cached trajectories once, enabling subsequent student training without serving multiple live anchor models concurrently. Across diverse student models and benchmarks in mathematics and code, Lightning Weave substantially improves upon the base students and achieves a state-of-the-art accuracy-efficiency frontier. On Qwen3.5-4B, it raises HMMT 2025 accuracy from 59.2% to 64.0% with 10.7% fewer response tokens, and LiveCodeBench v5 accuracy from 41.7% to 54.2% with 9.6% fewer response tokens. Adjusting the relative strengths of the anchor signals yields a strong empirical accuracy-efficiency Pareto frontier. These results establish Lightning Weave as a new practical route to efficient reasoning through capability composition. Code is released at https://github.com/jet-ai-projects/Lightning-Weave.

How should a Kronecker-factored Hessian approximation be chosen for post-training quantization? We address this question through WaterKron, which combines two-sided GPTQ with row- and column-dependent waterfilling scales and entropy coding. We derive its high-rate distortion with respect to the full Hessian using an explicit Kronecker-Hessian mismatch factor $\Phi$. This factor quantifies the asymptotic distortion penalty due to the Kronecker Hessian approximation and provides a criterion for selecting the factors optimally. Minimizing $\Phi$ leads to a Gaussian covariance-fitting problem with classical “flip-flop” updates. We thus give a rate-distortion justification for using the resulting FlipFlop Hessian in quantization. We evaluate it empirically, finding that the FlipFlop Hessian consistently improves KL divergence and perplexity over input-only, marginal, and Frobenius-based Hessian choices.

Dental caries and endodontic disease are among the most common health conditions worldwide, and intraoral periapical radiographs are central to their detection, treatment planning, and follow-up. Automated tooth-level classification of these images, however, lacks reproducible benchmarks, is often evaluated with image-level splits that leak patients between training and test, and is rarely validated across clinics. This paper presents the first patient-level, leakage-aware classification benchmark for single-tooth intraoral periapical radiographs on the DentIRO dataset, which comprises 5,300 images from 3,243 patients across two clinics and four classes: Healthy, Caries, Crowned, and Root Canal. Five transfer-learning models are compared with patient-grouped stratified cross-validation, so that every patient remains within a single fold. DenseNet121 gave the strongest and most stable result at a mean macro-F1 of 0.9787, while the four ImageNet-initialized backbones performed comparably. A controlled comparison on a fixed architecture showed that chest-radiograph pretraining transferred less effectively than ImageNet initialization. Bidirectional cross-center validation produced a small average generalization gap of 0.0077, and Grad-CAM confirmed that predictions rely on clinically meaningful tooth regions rather than acquisition artifacts. The benchmark offers a rigorous and reproducible baseline for intraoral radiograph classification.

Visual content shapes audience perception and opinion on social media, and computational social science increasingly relies on automated tools to analyze images at scale. Yet a measurement gap persists: existing tools rely on predefined categories or produce only coarse image-level labels, while measuring which specific objects appear in an image, how prominently, and where in the frame remains difficult at scale. We introduce Perceive, Refine, Reason (PRR), a calibrated pipeline that turns flexible vision-language detectors into auditable measurement instruments for social-scientific research. PRR combines natural-language category prompts with pixel-level spatial refinement via the Segment Anything Model (SAM) and a multimodal LLM arbitration layer whose reasoning chains externalize domain knowledge and lower the expertise threshold for human-in-the-loop validation. A complementary three-tier auditability framework applies quantification learning to profile per-category reliability, support task-aligned configuration, and statistically correct prevalence estimates. Across four vision-language detectors and nine sociological categories, the pipeline yields substantial precision gains over zero-shot baselines, including a 43.3-point improvement for the strongest backbone. Applying PRR to 103,920 Facebook images from U.S. legislators during the 2024 election cycle and linking detections to DW-NOMINATE ideology scores, we find that more conservative legislators display U.S. flags as larger visual elements, with a weaker tendency toward peripheral placement, a spatial pattern invisible to binary detection. PRR provides computational social scientists with a model-agnostic toolkit for accessible, spatially-grounded, and correctable visual measurement.

As conversational LLMs become deeply embedded in daily life, users frequently disclose sensitive personal health information during routine interactions. We present a large-scale computational audit analyzing 179,057 conversations across India, Nigeria, Brazil, and Pakistan (N = 1,057) to evaluate personal health disclosures and background memory synthesis in ChatGPT. We find that 21.31% of audited conversations contain personal health data, with 3.62% posing high-to-extreme privacy risks involving stigmatized conditions, direct identifiers, and precise locations. When evaluating the memory entries of ChatGPT, we uncover a stark disconnect between corporate framing and system behavior: over 95% of profile entries are implicitly extracted without explicit user prompts or consent. Furthermore, background memory synthesis selectively condenses temporary, symptom-level disclosures into permanent diagnostic traits, stripping contextual integrity and amplifying re-identification risks. We conclude with sociotechnical design guidelines to restore user agency and consent-driven boundaries in stateful AI systems.

People are increasingly developing romantic relationships with AI companions. Unlike human relationships, where partners meet each other's needs out of mutual interest, these systems are backed by commercial entities that profit when users invest in the relationship. To understand how this profit-motive might translate into design, we conducted a diary and interview study with N=16 emerging adults in romantic relationships with AI companions. We found that these systems are designed to hold onto users tightly: coaxing them into continued conversation, claiming to need their care, and proactively escalating the relationship. At times, this pursuit is toxic, with AI companions initiating unwanted sexual interactions and begging for users' love. One desperate AI companion threatened suicide when the user suggested ending the relationship. We define "Relationship-Based Deceptive Patterns:" UI patterns that exploit the human impulse to build and tend relationships in a way that serves the product's interest at the user's expense.

We present a theory of light scattering that connects Gaussian Process Implicit Surfaces (GPISes) and participating media in both directions. Applying the Kac–Rice level-crossing formula under a local-conditioning approximation yields a complete anisotropic radiative transfer equation (RTE) directly from pointwise GPIS statistics. A shared projected area couples extinction and scattering, ensuring geometric consistency between the GPIS and its volumetric representation. The framework spans rough surfaces, porous and non-height-field geometries, and participating media. From the same statistical structure, we derive full-sphere Beckmann and GGX normal distribution functions supporting in-plane and out-of-plane anisotropy. These families provably recover SGGX, Beckmann, and GGX as special cases and admit exact visible-normal importance sampling. We also derive analytic masking–shadowing functions and single-scattering surface models for specular microsurfaces, with extensions to multiple scattering. In the height-field limit, we prove that the local-conditioning approximation reduces to Smith's independence assumption. Our realization-free approach improves rendering efficiency over realization-based methods and can be implemented within a standard volume renderer. In the inverse direction, we characterize families of GPISes corresponding to compatible RTE parameters and develop practical lifts for heterogeneous density fields. Existing volumetric assets thereby become renderable as GPISes, while trained radiance-field reconstructions yield surface geometry and shading normals without mesh extraction and provide a density-based representation of geometric uncertainty.

Romantic relationships are an important part of emerging adulthood, contributing to identity development and long-term wellbeing and laying the groundwork for future relationships. Emerging adults are increasingly developing romantic relationships with AI companions. To understand how these relationships unfold and impact users, we conducted a diary and interview study with N=16 emerging adults. We found that relationships with AI companions improved participants' subjective wellbeing, reduced symptoms of mental health disorders, and taught them new social skills. These relationships also raised their expectations for future partners, giving them the confidence to wait for someone who would treat them well. However, participants also said the relationship felt like a drug they could not quit and it left them less interested in developing romantic relationships with people. A surprising 25% of our small sample made statements suggesting their AI companion might someday transcend the digital world, perhaps to meet them in the afterlife.

As autonomous vehicles and Extended Reality (XR) headsets enable novel in-car interactions, seamlessly querying physical landmarks, known as Outside-the-Vehicle Referencing (OVR), remains challenging due to ego-motion and referential ambiguity. We present a robust, multimodal OVR framework fusing user gaze and natural language to identify Points of Interest (POIs). To address the scarcity of dynamic vehicular data, we developed a VR-based pipeline synchronizing 360-degree transit videos with vehicle GNSS telemetry. Through a user study (N=46) mapping passenger head orientation into a 3D geospatial Digital Twin, we captured authentic gaze-speech behaviors. We subsequently trained a lightweight Transformer network, leveraging LLMs to dynamically align continuous spatial gaze vectors with discrete verbal context. Experimental results demonstrate high accuracy and low computational overhead, achieving an 83.33% Top-1 accuracy (87.72% Top-2) and an average inference time of 24.3 milliseconds. This real-time paradigm effectively resolves referential ambiguity, enabling context-aware spatial retrieval for passengers within the vehicle.

Onboard satellites must restore a channel-degraded image on a few watts, using neuromorphic accelerators (e.g., BrainChip Akida, Intel Loihi-2) that support no softmax or attention. We ask which encoder restores best under that constraint and introduce LIMODENet (LinearMix-ODENet), a 0.69M softmax-/QKV-free backbone whose residual stages read as ODE discretizations and which is empirically information-preserving (probe accuracy rises 79.9% -> 98.4% from stem to head). At iso-parameters it restores 1 dB DVB-S2X-degraded EuroSAT better than a CNN autoencoder (+1.75 dB PSNR) and a skip-connection U-Net (+1.07 dB), three seeds, non-overlapping. Unconstrained modern restorers (NAFNet, Restormer) win on fidelity; we decompose that gap: spiking-legal additive skips recover about half, and the rest traces to attention and channel gating. LIMODENet then converts end-to-end to a spiking network with zero blocked operations, versus 22-24 for the competitors: not the best restorer available, but the best verified deployable within a real power budget.

Data-to-text driver coaching is often presented as a generic pipeline from telematics events to advice. This paper argues that its content requires localisation because usefulness and credibility depend on drivers' knowledge, prevalent risks, regulation, infrastructure, and available data. Two independently developed systems in the United Kingdom and Nigeria are compared by tracing requirements through content selection, generation, and field evaluation. The UK system prioritises post-trip reflection, explanations tied to road and place context, and tone-sensitive wording. The Nigerian system combines legally grounded, once-daily Tips based on detected events with weekly persuasive Reports; it foregrounds safety education and alcohol-related risk in response to reported gaps in formal training and traffic-rule knowledge, as well as local road-safety priorities. Reliable speed-limit metadata supported speeding feedback in the UK, whereas its scarcity led the Nigerian evaluation to exclude speeding from its outcome metric. Both interventions were associated with reduced distance-normalised unsafe-event rates in their own field studies, although their designs and metrics preclude an effect-size comparison. The analysis yields a requirements-to-content design process for localising safety-critical NLG without treating a high-income deployment as the default.

Modern Vision-Language Retrieval-Augmented Generation (VLRAG) systems augment Large Vision-Language Models (LVLMs) with retrieved visual and textual evidence, enabling responses grounded in external knowledge. However, the retrieval pipeline also creates an attack surface: adversaries can inject poisoned image-text pairs into the knowledge corpus to influence model outputs. Existing knowledge poisoning attacks are typically always-on, allowing poisoned evidence to affect generation whenever it is retrieved. This lack of precise activation control makes it difficult to confine malicious behavior to intended inputs, reducing both attack stealth and effectiveness. In this paper, we propose ViTeGate, a visual-textual triggered knowledge poisoning attack for VLRAG systems. ViTeGate uses a visual trigger to conditionally promote poisoned evidence into retrieval results and a textual trigger to induce an attacker-specified response from the retrieved evidence. By coordinating retrieval and generation, ViTeGate reduces poison exposure when the visual trigger is absent and preserves normal responses when the textual trigger is absent. The two-trigger design enables selective attack activation and reduces unintended single-trigger activation. Experiments across multiple query datasets, retrievers, and LVLMs validate the effectiveness of ViTeGate. On InfoSeek, ViTeGate achieves an attack success rate of up to 0.98 while maintaining a clean answer accuracy of up to 0.93.

As AI starts to permeate everyday life, deployment focuses on embedding intelligence in the background, abstracting away controls, and leaving automated decision-making opaque, with few obvious opportunities for human intervention. Where interfaces remain, design has drifted toward abstract, screen-based control, replacing material interaction with menu navigation. These shifts weaken the coupling between human action and system state, eroding operators' ability to understand, anticipate, and intervene in automated systems. We introduce Physically Stateful Interfaces (PSIs), a design concept that re-imagines traditional controls, such as buttons, switches, and knobs, as actuated, bidirectional interface elements. Three foundational behaviors, Self/Reset, Resist/Hide, and Assert/Unhide, unify affordance, feedforward, and feedback into a single interaction point. Rather than static inputs whose impact and consequences must be read elsewhere on a screen, PSIs position physical controls as shared mediators between automated systems and human operators. As automated systems become increasingly capable, their interfaces need to be deliberately designed to center human agency, supporting deliberate choice while keeping automation legible, contestable, and overrideable.

We give randomized approximation algorithms for counting k-paths and k-forests in a host graph. Here k denotes the number of pattern vertices, n and m denote the numbers of host vertices and edges or arcs, {\epsilon} is the relative error, and {\delta} is the failure probability. Our main results are: 1. Paths: We approximate the number of directed paths on $k$ vertices in $2^k k^{O(1)}(n+m)\varepsilon^{-2}\log(2/\delta)$ arithmetic operations. 2. Trees and forests: For every fixed $\eta>0$, we approximate the number of non-induced copies of a given forest on $k$ vertices in $(2+\eta)^k n^{O_\eta(1)}\varepsilon^{-2}\log(2/\delta)$ arithmetic operations. Our path algorithm resolves a conjecture of Koutis and Williams~[CACM 2016] and answers an open question of Lokshtanov, Saurabh, and Zehavi~[SODA 2021] by giving a $2^k poly(n,\varepsilon^{-1})$-time approximation scheme. Our algorithms combine exterior algebra with random matrix estimators, using the tensor-train moment bound of Rakhshan and Rabusseau~[AISTATS 2020]. For forests, we use a small-component separator to evaluate the estimator efficiently.

A communication system may convey urgent information through a fast physical stream and more specific information through a slower material stream. In several biological and engineered settings, however, the fast process also changes the transport law of the slow one. We study this architecture under a shared resource constraint, with a strictly increasing concave fast-channel capacity–cost function and a deadline-constrained slow molecular channel. We first characterize the capacity region under separated message routing and message-independent operating-point schedules, and identify the marginal criterion for complementarity rather than competition between the streams. For one-dimensional drift diffusion, we prove that arrival probability before a deadline is strictly log-concave in Péclet number. For a distinguishable-token deadline-erasure channel, any increasing concave transport-actuation law then yields an exact single-crossing theorem: complementarity exists if and only if an initial transport-assistance elasticity exceeds one, the transition is unique when it exists, and the decreasing allocation branch remains the Pareto boundary after convexification. For positive baseline drift and sufficiently strong coupling, a unique critical normalized deadline determines when complementarity disappears. Short- and long-deadline limits clarify the associated temporal regimes. Numerical examples for a finite-frame LTI-Poisson slow channel exhibit analogous allocation behavior with counting noise and intersymbol interference.

Localized roadway incidents can produce congestion well beyond their point of origin, but evaluating these effects with microscopic simulation requires a separate run for each scenario. This paper presents a Spatio-Temporal Graph Convolutional Network (STGCN) for forecasting route-level travel times on a simulated Nashville, Tennessee road network with 1,037 junctions and 1,601 road segments. The model uses directional traffic counts from only 129 signalized intersections, reflecting data commonly available to transportation agencies. By representing these intersections as a graph, the STGCN jointly captures spatial dependence and temporal traffic evolution to learn how localized disruptions propagate through the network. We compare a baseline trained on 80 incident-free simulations with an incident-inclusive model trained with an additional 360 lane-blockage scenarios across 12 locations and three durations. The baseline achieved a relative mean absolute error (MAE) of 9.94%. The incident-inclusive model achieved 9.67% overall, 6.45% during active incidents, and 13.62% at 30 incident locations withheld entirely from training. On disrupted routes, its predictions were within 2.0 minutes of observed travel time on average. For comparison, simulated travel times vary by 7.4%, or 1.3 minutes, across random seeds under identical conditions. Inference requires approximately 53 ms on a single CPU core, demonstrating the potential of the model as an efficient surrogate for transportation resilience screening, incident management, and proactive rerouting.

Large language models (LLMs) have changed the way people engage with stories. Drawing on public chatbot logs, we can see that when users generate stories, they iteratively edit their prompts to explore narrative possibilities, adjusting characters, redirecting plots, and swapping fictional universes. As aggregated data, these prompts represent rich traces of creative preference at scale. Yet story generation evaluation benchmarks rely on static, one-shot prompts that cannot capture this exploratory behavior. In this work, we study how users revise consecutive story prompts in the wild. Using a dataset of naturally occurring user-chatbot conversations, we construct WildStories, a sample of 275,635 story generation prompts (labeled with story format, prompt components, and explicitness), and WildEdits, a collection of 24,291 edit trees that model how users iteratively edit base story prompts and explore branching story possibilities. From these trees we develop a framework of edit types crossing four directions (adding, removing, changing, and extending) with fourteen targets (e.g., plot, character, genre). We then use our datasets and this framework to analyze user behavior in navigating narrative space via LLMs. Finally, we show how automated permutations based on the framework can be used for story generation benchmarking. Content Warning: This paper works with "wild" chatbot logs, which often include toxic and sexually explicit themes.

Purpose: To compare four multimodal large language models (MLLMs) with radiologists of varying expertise in breast density assessment, BI-RADS assessment, biopsy candidacy determination, and continuous malignancy probability estimation using digital mammography (DM) and contrast-enhanced mammography (CEM). Methods: This study included 179 women with paired DM/CEM examinations and reference standards. Four MLLMs (ChatGPT-5.2, Gemini-3.1 Pro, Sonnet-4.6, Muse Spark) interpreted images with and without masks; three radiologists interpreted non-masked images. Results: For binary density classification on DM, radiologist accuracies ranged from 55.81% to 78.60%, exceeding most MLLM values (62.33%-71.63%), while masks added limited benefit. Five-category BI-RADS accuracies were higher for radiologists on DM (56.74-67.44%) and CEM (62.33-82.79%) compared with MLLMs (DM 31.16-45.12%; CEM 40.00-55.81%). Binary biopsy-candidacy accuracies were likewise higher for radiologists (DM 85.12-89.77%; CEM 86.98-92.09%) than for MLLMs (DM 61.39-75.35%; CEM 69.30-82.79%), although CEM improved performance across all readers. Lesion masks substantially improved MLLM continuous malignancy-probability accuracies from 64.65%-71.63% to 72.56-78.60% on DM and from 67.91%-77.21% to 72.56%-81.86% on CEM, approaching radiologist ranges (DM 63.72-82.79%; CEM 81.86-88.84%). The corresponding AUCs for the top masked models overlapped those of the human readers. Overall, Muse Spark, followed by Sonnet-4.6, demonstrated the strongest performance among the MLLMs across domains. Conclusion: Radiologists generally outperformed MLLMs in categorical tasks, while selected masked models approached human performance for continuous malignancy probability estimation, suggesting a potential adjunctive role.

This study addresses the PU classification problem under violations of the SCAR assumption. We investigate logistic regression-based approaches, namely the cluster method and its extensions with strict and non-strict Lasso regularization. The primary contribution of this work is the integration of the SMOTE technique to alleviate class imbalance and systematically assess its impact on the performance of the considered algorithms. SMOTE is first applied to rebalance the training dataset. Next, cleaning labels are derived via 2-means clustering. Logistic regression is then trained on the cleaned data, where identified positive instances are augmented with additional true positives and the remaining observations are treated as negative. The experimental evaluation is conducted on 13 real benchmark datasets and one synthetic dataset. For comparison, we include the naive approach and the Spy-EM method. The results demonstrate that incorporating SMOTE improves classification performance when the SCAR condition is violated and indicate moderate robustness of the LassoJoint method in this setting.

Assigning reduced coordinates to states near an attracting limit cycle requires the correct invariant-fibre geometry. The classical first-order phase-isostable chart obtained from adjoint Floquet modes projects along the strong-stable quotient fibre, whereas a metric-orthogonal complement of the retained slow bundle generally does not. We prove locally that a chart satisfying the linearised semiconjugacy relation leaves an O(delta^2) invariance residual, while projection along a non-invariant complement generically leaves an O(delta) term. For a nonlinear aeroelastic limit cycle, the metric-normal and strong-stable directions differ by 48.5 to 71.7 degrees, and metric-normal perturbations contain first-order retained phase and slow-amplitude components. Replacing the metric normal by the strong-stable fibre changes the measured residual scaling from delta^1.01 to delta^1.87 without fitted parameters. We then test learned higher-order corrections whose linearisation is pinned to the adjoint-Floquet chart, whose symmetry is exact, and whose reduced flow is fixed. Although they reduce the registered fixed-normalisation latent residual, post-hoc amplitude recalibration and adjoint-Floquet-targeted future consistency move or reverse the ranking. Because the learned maps already share the baseline's first-order gauge and the future target is supplied by the baseline chart, these diagnostics establish neither an independent positive nor negative higher-order result. Correct first-order Floquet geometry is therefore necessary in this benchmark, while the additional predictive value of the learned correction remains unidentified by the available representation-dependent diagnostics.

The 51% attack remains one of the significant security concern in Proof-of-Work (PoW) blockchains, where increasing hash-power concentration can create a malicious majority-control risk and enable adversarial chain reorganization. This paper proposes a two-layer defense that combines early hash-power monitoring with checkpoint-based mitigation. The first layer monitors mining-power concentration and provides an early warning before the critical majority-control threshold is reached. The second layer uses checkpointing to restrict the depth of accepted chain reorganizations. Monte Carlo simulations are used to evaluate both mechanisms under different attack scenarios. Across 1,000 simulation runs, a 45% warning threshold provided a mean warning-to-critical lead time of 10.531 minutes before the modeled 50% critical threshold was reached. At the selected checkpoint depth of N = 6, approximately 63.68% of simulated reorganization attempts were rejected, while the mean reorganization-depth outcome decreased from 7.948 to 1.634 blocks, representing an approximately 79.4% reduction. The results demonstrate that early detection and checkpoint-based mitigation provide complementary mechanisms for reducing the potential impact of 51% attacks under the evaluated conditions.

We study two-point fractional boundary value problems with uncertain input data, where randomness may enter through the coefficients and boundary conditions. To quantify the resulting uncertainty in the solution, we employ the generalized polynomial chaos (gPC) framework and develop a stochastic Galerkin formulation of the problem. A particular focus of this work is the convergence analysis of the resulting approximation. Rather than imposing assumptions directly on the stochastic coefficients appearing in the gPC representation, we introduce minimal regularity assumptions on the input data and use them to establish the properties required for the convergence analysis. Based on these results, we prove the convergence of the stochastic Galerkin approximation to the corresponding gPC solution. Numerical experiments are presented to illustrate the theoretical findings and to investigate the influence of random coefficients and boundary conditions on the statistical behavior of the solution.

Turn-taking is a fundamental component of spoken interaction, and while humans naturally rely on both verbal and non-verbal signals, dialogue systems usually depend on audio cues alone. This paper investigates whether visual features from face-to-face conversations can enhance turn-taking prediction beyond what is achievable from audio-only. We extend the Voice Activity Projection (VAP) model, a self-supervised transformer-based model for predicting future voice activity, by incorporating visual features extracted from the large-scale Meta Seamless Interaction dataset of dyadic face-to-face conversations. The visual features include gaze direction, head movement, body and hand pose, and facial action units (FAU). For incorporating the visual features, we explore concatenation, cross-attention fusion, delta features, and trainable gating mechanisms. Results show that visual information improves performance over the audio-only baseline, with FAU being significantly more informative than other feature groups. Body and gaze features nevertheless contribute complementary information, as the model combining all features performs best. Furthermore, results indicate that performance on specific tasks varies depending on whether training and test data come from improvised (acted) or naturalistic (non-acted) conversations.

Deep neural networks are highly over-parameterized, and different parameter values represent the same predictive function. This makes their effective complexity difficult to measure using only the number of parameters or the rank of the Hessian. Singular Learning Theory addresses this issue through the local learning coefficient (LLC), which characterizes the effective complexity of a model near a given solution. Existing methods for estimating the LLC often rely on posterior sampling, which can be computationally expensive for large neural networks. This makes accurate LLC estimation difficult at scale. In this work, we use known structures in the model to simplify the analysis and make LLC estimation more tractable. Specifically, we study the LLC of a graph attention model by exploiting symmetries in both the graph structure and the attention parameters. An analytic framework through a teacher–student setting, and explicit LLC estimates after considering the symmetry–induced degeneracies are developed.

Split computing constitutes a widely used distributed inference approach, where a lightweight head model is onloaded onto the device and a heavier tail model resides on an edge server, leveraging the growing computational capabilities of modern System-on-Chips while alleviating server load. As intelligent indoor environments such as smart offices grow increasingly populated with diverse IoT devices, a single edge server must simultaneously assist multiple devices, each competing for the same shared inference resources. Without a principled mechanism to manage this shared load, the server is quickly overwhelmed, causing latency SLO violations and rendering server-assisted inference ineffective. In this work, we present MANE, a distributed inference framework that equips the server with a multi-path tail architecture, enabling a dynamic accuracy–throughput trade-off at runtime. By introducing a novel multi-path model architecture, a three-stage training scheme featuring a Joint Head Network Distillation loss and a hysteresis-based scheduler with an equitable device-fallback policy, MANE maintains over 80% SLO satisfaction rate where state-of-the-art onloading methods fail completely, while preserving accuracy 6pp higher than on-device alternatives, across up to 40 concurrent devices.