arXiv:2607.00403v1 Announce Type: new Abstract: Large language models are increasingly used by participants on crowdsourcing platforms when responding to surveys, potentially undermining the validity of collected data. Our study aims to quantify the prevalence of this behavior and investigate methods to detect and prevent it. In a series of surveys (N = 250), we examined conditions such as platform choice, survey length, requests not to use AI, and disabling copy-paste functionality. We were able to identify distinct characteristics of LLM-assisted responses and found that their frequency varied widely, from under 10% on Prolific to over 80% on Mechanical Turk. Mitigation measures reduced LLM usage but did not necessarily improve data quality. No participants employed browser-use agents at the time of our survey, but we report on our own detection experiments. We recommend that researchers actively screen survey responses for LLM usage by recording and analyzing keystroke data and crafting instructions and questions aimed at AI.
Science Journals
arXiv:2606.08492v2 Announce Type: replace Abstract: Despite the impressive capabilities of text-to-image (T2I) models, an intent-generation gap often persists due to the brevity and ambiguity of user prompts. Existing approaches primarily polish the prompt for fluency and readability. However, the enhancement process still lacks visual grounding. As a result, the rewriter may over-infer missing details, causing an intent-generation gap. To address this limitation, we propose FaithRewriter, a novel prompt-enhancement framework for T2I generation. Specifically, FaithRewriter first leverages a multimodal MLLM to generate an image from the original prompt as an intermediate visual cue. This cue is then combined with the prompt and fed into a large-scale LLM to produce visually grounded augmentations that better reflect how the intended content should appear in images. Finally, these augmentations are distilled into a small-scale LLM for efficient deployment, enhancing its ability to generate effective T2I prompts. Experiments show that FaithRewriter yields prompts that are more faithful to the user intent and more visually plausible than strong baselines, helping narrow the intent-generation gap.
arXiv:2606.08625v2 Announce Type: replace Abstract: As Large Language Models (LLMs) advance toward open-ended autonomous agents, the mechanisms used to evaluate and guide their behavior must evolve accordingly. This work introduces the rubric as a unifying framework capturing this evolution, characterizing rubrics as a dynamic response to successive LLM paradigm shifts that recurs across otherwise independent efforts in evaluation, reinforcement learning, and safety alignment. We define rubrics as explicit criteria sets that transform complex quality judgments into structured and actionable standards, and demonstrate that their recurrence across these research threads is not coincidental. We systematically organize existing rubric designs, examine their construction and optimization, and analyze their role across evaluation and training. Rubrics manifest at three progressively deeper levels: at the evaluative level, they decompose holistic judgments into verifiable dimensions; at the training level, they serve as dense feedback signals providing process-level guidance where scalar rewards fall short; at the intrinsic level, they emerge dynamically from model behaviors, driving self-improvement. We further assess rubric reliability across generation quality, execution fidelity, theoretical constraints, and security threats, before surveying rubric-based benchmarks across diverse domains. By rendering assessment transparent and decomposable, rubrics translate human value expectations into machine-learnable signals, serving as the enduring bridge between human intentions and machine behavior.
arXiv:2606.09606v3 Announce Type: replace Abstract: Ray tracing enables 3D Gaussian fields to serve as a representation for physically based light transport. Faithful inverse rendering requires forward rendering and backward optimization to be defined within a consistent light-transport pipeline. Existing inverse rendering methods estimate G-buffers via splatting and optimize materials in screen space, tying the recovered properties to a rasterization-based pipeline. This pipeline mismatch, together with simplified rendering equations that neglect indirect illumination, often leads to inconsistent shading, visible artifacts, and inaccurate material-lighting estimation under path-traced rendering. Therefore, we propose a splatting-free path-traced inverse rendering framework for 3D Gaussian fields, where forward light transport and backward gradient propagation are defined within a unified ray-tracing pipeline. Our key idea is to define a path-space equivalent interaction model for overlapping Gaussian primitives, under which Monte-Carlo-based path tracing is unbiased for the induced light-transport integral, while pathwise gradients are replayed over the same ray-traced interactions rather than splatting-derived screen-space buffers. The framework optimizes materials and a compact Spherical-Gaussian environment under the full rendering equation with ray-traced visibility and multi-bounce light transport. Extensive experiments demonstrate competitive material inversion and improved path-traced rendering quality, producing more plausible shadows, reflections, and relighting results under global illumination.
arXiv:2606.10111v2 Announce Type: replace Abstract: This paper presents a nonlinear parameter estimator for Wiener-type state-space models obtained as a fixed-point architecture that couples two affine minimum mean-squared error (MMSE) estimators: one for the unknown parameters and one for latent variables. The architecture retains the functional structure of the optimal affine MMSE parameter estimator while incorporating Dynamic Basis Statistics (DBS) estimates that summarize nonlinear basis-function evaluations. Two DBS construction strategies are developed, leading to two nonlinear estimator frameworks. The dual basis-parameter estimator combines an affine basis estimator with the affine parameter estimator, whereas the dual state-parameter estimator first computes affine state estimates and their covariances, then maps these state-estimate statistics through a Gaussian DBS operator to obtain DBS estimates. Both dual estimators admit fixed-point characterizations that alternate between estimating each component using the updated prior of the other, obtained from that component's plug-in estimate statistics from the previous iteration. The efficacy of the proposed methods is examined via extensive Monte Carlo experiments, showing that the dual basis-parameter estimator attains parameter mean-squared errors comparable to those of the purely affine parameter estimator, while the dual state-parameter estimator achieves the lowest parameter mean-squared error, outperforming both the dual basis-parameter and purely affine parameter estimators, as well as sequential Monte Carlo variants of classical Particle Gibbs and Expectation-Maximization schemes.
arXiv:2606.10531v2 Announce Type: replace Abstract: Quantization-aware training (QAT) is essential for extremely low-bit large language models (LLMs). Current QAT methods are mainly based on scalar quantization (SQ), which enables efficient optimization but suffers from severe performance degradation at 2-bit precision. On the other hand, vector quantization (VQ) provides substantially higher representational capacity, but its discrete codebook lookup prevents end-to-end training. We propose LC-QAT, a 2-bit weight-only VQ-QAT framework that represents quantized weights via a learned affine mapping over discrete vectors, which yields a high-quality PTQ initialization and enables fully differentiable end-to-end optimization without explicit codebook lookup in the training forward pass. This strong post-training initialization makes LC-QAT highly data-efficient. Experiments across diverse LLMs demonstrate that LC-QAT consistently outperforms state-of-the-art QAT methods while using only 0.1%--10% of the training data. Our results establish LC-QAT as a practical and scalable solution for extreme low-bit model deployment. Codes are publicly available at https://github.com/AI9Stars/UniSVQ.
arXiv:2602.23164v2 Announce Type: replace Abstract: Foundation models must handle multiple generative processes, yet mechanistic interpretability largely studies capabilities in isolation; it remains unclear how a single transformer organizes multiple, potentially conflicting "world models". Previous experiments on Othello playing neural-networks test world-model learning but focus on a single game with a single set of rules. We introduce MetaOthello, a controlled suite of Othello variants with shared syntax but different rules or tokenizations, and train small GPTs on mixed-variant data to study how multiple world models are organized in a shared representation space. We find that transformers trained on mixed-game data do not partition their capacity into isolated sub-models; instead, they converge on a mostly shared board-state representation that transfers causally across variants. Linear probes trained on one variant can intervene on another's internal state with effectiveness approaching that of matched probes. For isomorphic games with token remapping, representations are equivalent up to a single orthogonal rotation that generalizes across layers. When rules partially overlap, early layers maintain game-agnostic representations while a middle layer identifies game identity, and later layers specialize. MetaOthello offers a path toward understanding not just whether transformers learn world models, but how they organize many at once.
arXiv:2606.11861v2 Announce Type: replace Abstract: We report ground-based coherent VHF radar observations of extreme turbulent field-structures detected in coincidence with a magnetospheric substorm-associated magnetotail dipolarization. The field-structures are observed by the ICEBEAR radar, in the form of Farley-Buneman (FB) waves in the auroral electrojets, and the field-structures themselves move an order of magnitude faster than the saturation speed of the underlying FB waves, implying transient electric field sources up to 330 mV/m in strength. The field-structures are identified and automatically tracked using an unsupervised clustering & tracking algorithm, applied to clutters of ICEBEAR radar backscatter targets, a method that turns the Doppler radar into a tracking radar capable of measuring the ionospheric ExB-drift by proxy. We place this finding in a coordinated multi-instrument context. Three THEMIS spacecraft observed the dipolarization event in-situ in the near-Earth plasma sheet. In the ionosphere, Swarm A, crossing through the guilty auroral arc at the onset of the dipolarization event, recorded clear signatures of propagating Alfv\'en waves threading the relevant flux tube. We interpret the ICEBEAR transients as the natural ionospheric foot signature of a shear Alfv\'en pulse launched by the bipolar space-charge (Hall) electric field of the thinned current sheet, with amplification along the converging flux tube, partial reflection at the ionospheric boundary, and spatial sharpening by precipitation-produced Pedersen-conductance gradients on the auroral arc edges. A one-dimensional wave-transmission analysis recovers the observations. Our results elucidate a tightly controlled coupling between magnetotail processes and meter-scale auroral plasma turbulence, and demonstrate the capability of ICEBEAR to resolve extreme, transient electric-field enhancements in the ionosphere.
arXiv:2606.13148v2 Announce Type: replace Abstract: Climate and environmental decision-making increasingly requires reasoning across heterogeneous inputs, including gridded physical data, satellite imagery, geospatial context, and simulator outputs. Weather and climate foundation models can forecast well, but do not reason interactively in language, while large language models (LLMs) reason in language but cannot operate directly on high-dimensional Earth-system data. As a result, real scientific workflows in Earth-science remain underserved. We introduce TerraBench, a benchmark for grounded Earth-science reasoning, built on TerraAgent, a ReAct-style executable framework that interleaves reasoning, tool calls, and observations to couple LLM planning with scientific tools for environmental retrieval, geospatial processing, simulation, and artifact-backed computation. TerraBench unifies analysis of Earth observation imagery, gridded data, GIS reasoning and simulation in a single executable interface, whereas prior benchmarks isolate these capabilities into narrow individual tasks. It is also the first in this space to pair process-level tool-use metrics with tolerance-aware numeric scoring. The benchmark comprises 403 extensive agentic tasks across three tracks (Fundamentals, Simulator-Grounded, and Document-Grounded Verification) and eight application domains with 24,500 verified execution steps. These results indicate that reliable Earth-science agents must go beyond tool access to coordinate heterogeneous workflows, parameterize tools precisely, and preserve artifact provenance.
arXiv:2606.13695v2 Announce Type: replace Abstract: Mineral prospectivity modelling (MPM) underpins exploration economics, yet most operational pipelines reduce to data-driven classifiers trained on shallow surface proxies. Such models are blind to the subsurface physics that actually localises ore: heat advection, fluid flow, and lithology-dependent precipitation. We present Korzhinskii-Net, a 2-D radial physics-informed neural network (PINN) that couples Darcy flow, advective-diffusive heat transport, and a softplus-saturated reaction rate into a single differentiable forward model, weakly supervised by surface and remote-sensing proxies. The network is named after Dmitri S. Korzhinskii (1899-1985), whose theory of infiltration metasomatism provides the physical scaffold. We evaluate Korzhinskii-Net on six ore provinces spanning three commodity classes - Udokan (sandstone-hosted Cu), Sukhoi Log, Olimpiada, and Berezovskoye (orogenic Au), Vorontsovskoye (Carlin-type Au), and Dalnegorsk (skarn polymetallic) - under a fair, leakage-controlled 5-fold cross-validation protocol with hard ring-shaped negatives and baseline proxy features disabled. Korzhinskii-Net attains a mean PR-AUC of 0.708 versus 0.235 for the strongest classical baseline (support vector machine), and a mean fractional rank of 0.036 versus 0.475. The improvement is consistent across all six provinces and three commodity systems, suggesting that physics-informed differentiable simulators, even when constrained only by global open-data proxies, can recover localisation patterns that pure feature-based learners systematically miss. We release the full pipeline and evaluation harness as open source.
arXiv:2606.13847v2 Announce Type: replace Abstract: Hyperscale AI data centers induce spatially and temporally correlated load fluctuations that violate classical independence assumptions and are not captured by time-averaged spectral methods. These correlations are episodic and non-stationary, so they demand analysis that resolves transient structure. This paper applies Dynamic Mode Decomposition (DMD) to the temporal evolution of pairwise inter-bus correlation coefficients and forms a low-dimensional state representation that enables modal analysis without a stationarity assumption. The recovered modes distinguish sustained coherence, decaying transients, and intensifying events, and their oscillation timescales map to underlying physical coupling mechanisms. The method is evaluated on an IEEE 39-bus Real-Time Digital Simulator (RTDS) testbed with three converter-interfaced AI data center loads driven by synthetic workload profiles. A global analysis attributes the dominant correlation energy to a slow thermal band, and a sliding-window analysis identifies brief intensification events in a small fraction of windows that align with stochastic workload coincidences. Cross-validation with RTDS voltage coherence confirms elevated coupling during these intervals. The proposed modal growth indicator provides an early-warning signal of correlation intensification, with a lead of of about 4~s before pairwise coherence reaches its peak.
arXiv:2606.13971v2 Announce Type: replace Abstract: While personalizing Image-to-Video (I2V) diffusion models with specific visual effects is increasingly demanded for high-end generation, current practice requires training a separate Low-Rank Adaptation (LoRA) module for each effect, incurring substantial data curation and iterative optimization costs that hinder interactive control. We present Prompt2Effect, a weight-driven hypernetwork that amortizes per-effect training by directly synthesizing effect-specific LoRA weights in a single forward pass. Unlike prior hypernetworks that regress adapter weights purely from semantics, Prompt2Effect is explicitly conditioned on the frozen base model weights, grounding prediction in the structural geometry of each layer. Furthermore, instead of predicting raw LoRA matrices, we introduce an SVD-canonicalized parameterization that resolves factorization ambiguity and stabilizes large-scale synthesis. Extensive experiments demonstrate that Prompt2Effect achieves on-par or superior video quality and effect alignment compared to conventional LoRA fine-tuning, while reducing the computational cost from 56 GPU training hours to 3.3 seconds of hypernetwork inference. When used as initialization for subsequent fine-tuning, our predicted weights further improve final performance and accelerate optimization by approximately 10x.
arXiv:2607.00509v1 Announce Type: new Abstract: Positron emission tomography (PET) provides essential functional information for disease assessment, however reducing injected activity or acquisition time produces low-dose (LD) PET with stronger count dependent noise and less reliable uptake quantification. Diffusion models offer a promising solution for PET denoising by progressively recovering high-dose (HD) PET images from LD inputs. However, LD-to-HD PET denoising is still challenging due to insufficient anatomical guidance, unstable multi-scale feature propagation, and uncertain frequency domain uptake recovery. We propose AnF-DiffPET, an anatomy- and frequency-guided diffusion framework for computed tomography (CT) conditioned LD PET denoising. The framework integrates Anatomical-Frequency Guidance (AFG), Multi-Scale Cross-Transformer Reconstruction (MSCTR), and Frequency-Contrastive Hard Mining (FCHM) to enhance anatomy aware feature modulation and frequency domain consistency during denoising. Experimental results across four PET/CT datasets show that the proposed method improves image fidelity, anatomical consistency, and quantitative fidelity over representative CNN-based, GAN-based, transformer-based, and diffusion-based methods. The code and trained models will be publicly released upon acceptance.
arXiv:2607.00518v1 Announce Type: new Abstract: Conventional Mixed Reality (MR) workspaces are frequently organized in cockpit-like layouts, where multiple floating windows surround the user. While this configuration facilitates access to digital content, it often induces occlusion, reducing understanding of the physical environment and limiting access to real-world objects. To overcome this challenge, we present the Contour-Adaptive Mixed Environment Overlays (CAMEO), a contour-adaptive MR interface that drapes virtual windows onto physical surfaces. This design integrates digital content with nearby items, thereby improving users' visual access to background objects and supporting interaction with them. We evaluate CAMEO in two controlled studies. The first demonstrates that draping reduces hand-movement detours relative to flat mid-air surfaces, enabling more direct interaction with nearby items. The second shows that controlled window deformation does not significantly impair text legibility when compared to flat surfaces. Together, these findings contribute a novel design paradigm for MR workspaces that balances immersion, readability, and environmental understanding.
arXiv:2606.14611v2 Announce Type: replace Abstract: Interdiction problems arise in a number of application areas, including global security, supply chains, and critical infrastructure protection - the goal is inhibit the movement of goods, people or information. An area of particular interest is counter-narcotics, where nodes or edges in a network are placed under surveillance or blocked to minimize the flow of illicit drugs from source to the destination. A fundamental challenge in this narco-traffic interdiction is data scarcity: available datasets are limited by the very nature of the problem and provide only partial and uncertain views of trafficking networks. Thus, developing robust interdiction methods that take this inherent lack of information is critical. In this paper we initiate the study of network flow interdiction problems under network uncertainty. First, using a limited real-world dataset, we generate an ensemble of plausible network realizations representing alternative trafficking scenarios. The method combines simulations with mathematical programming techniques to generate network ensembles that are consistent with the observed data. Second, we formulate the robust network flow interdiction problem and develop an integer linear program to solve the problem. We evaluate the optimal interdiction strategy and obtain the residual flows over the scenarios. Our analysis reveals that even modest budgets can yield significant flow reductions. However, optimal solutions vary substantially across scenarios, motivating the need for robust solutions. We show that the robust strategy achieves near-optimal performance across all near-real world realizations while remaining stable under structural uncertainty. This simulation-driven approach provides a principled basis for policy analysis and supports maximizing the return on interdiction investments in uncertain, data-limited environments.
arXiv:2606.16511v2 Announce Type: replace Abstract: Recent work motivates moving large language model (LLM) evaluation from mean-based to tail-aware metrics, including conditional value-at-risk and tail-index estimates of reward-model error. We ask whether the canonical extreme-value-theory tail-index parameter, which isolates how heavy a tail is from how large the tail mass is, adds discriminative information beyond the mean and a standard tail-magnitude statistic in LLM evaluation. We pre-register a protocol covering admissibility, goodness-of-fit, threshold-stability, and effect-size requirements for any positive tail-shape claim. The protocol is the contribution of this paper; the empirical study below is a demonstration of what its gates catch. Applied to a standard LLM toxicity-evaluation setup under two structurally different scorer families, the protocol catches three distinct modes of false positives that a naive analysis would have published, and rejects the headline tail-shape claim on both scorers. We conclude that tail-shape estimation in the LLM toxicity-evaluation setups we examined is more fragile than the recent literature suggests, and recommend the protocol as a starting point for tail-index claims in similar setups.
arXiv:2606.17041v5 Announce Type: replace Abstract: Meta-analysis is a demanding form of evidence synthesis that combines literature retrieval, PI/ECO-guided study selection, and statistical aggregation. Its structured, verifiable workflow makes it an ideal substrate for evaluating systematic scientific reasoning, yet existing benchmarks lack ground truth across the full retrieval-screening-synthesis pipeline. We introduce MetaSyn, a dataset of 442 expert-curated meta-analyses from Nature Portfolio journals. Each entry pairs a research question with PI/ECO criteria, a retrieval corpus of 140k PubMed articles, verified positive studies, hard negatives that are topically similar but PI/ECO-ineligible, and complete search strategies and date bounds. Benchmarking twelve pipeline configurations (nine RAG variants and a protocol-driven agent) reveals a critical screening bottleneck: despite a retrieval ceiling of 90.9% recall at K=200, no system recovers more than 52.7% of ground-truth included literature. Current LLMs fail to reliably separate eligible studies from PI/ECO-failing distractors in pools of comparable topical relevance. Stage-attributed metrics capture where systems succeed and fail; a single end-to-end score does not.
arXiv:2606.17416v2 Announce Type: replace Abstract: Multilingual speaker verification remains challenging because language-dependent acoustic variability causes speaker identity to become entangled with linguistic characteristics, degrading generalization across languages. In multilingual training, embeddings often encode language cues with speaker identity, causing speakers to form language-specific clusters. We propose L-Proto, a language-aware episodic prototypical training strategy that constructs language-consistent episodes. By sampling speakers from a single language per episode, L-Proto reduces language-driven variation during training and encourages embeddings to focus more directly on speaker identity. Experiments on the TidyVoice Challenge benchmark demonstrate consistent performance improvements over conventional fine-tuning and random episodic sampling across multiple backbone architectures.
arXiv:2606.18367v2 Announce Type: replace Abstract: Standard benchmarks evaluate time series foundation models (TSFMs) using aggregate metrics, but these can mask severe failures in critical operating regimes. We introduce regime-stratified evaluation and apply it to three TSFMs on two standard traffic speed benchmarks. Traffic exhibits abrupt regime switching between free-flow and congested states, producing bimodal speed distributions during transitions. When we stratify by traffic regime, both accuracy and prediction-interval coverage degrade sharply during transitions: transition-regime MAE reaches 11 mph (versus 3 mph overall), and empirical coverage of 90% prediction intervals drops as low as 55%. These failures are invisible in aggregate metrics because free-flow observations dominate the sample. A simple historical conditional baseline (sampling from per-sensor training distributions) achieves better transition coverage than any TSFM, but has far worse overall accuracy. We propose bimodal mixture augmentation (BMA), a post-hoc method that combines TSFM forecasts with historical distributional knowledge, approaching the historical baseline's transition coverage while preserving the TSFM's accuracy. Our results suggest that TSFM benchmarks should incorporate regime-aware evaluation to surface failures that aggregate metrics hide.
arXiv:2606.19096v2 Announce Type: replace Abstract: European Portuguese (pt-PT) is largely absent from OCR benchmarks, which skew toward high-resource languages. The few benchmarks that cover pt-PT focus on historical artifacts and literature. This work addresses modern OCR applications, introducing PorTEXTO, the first benchmark for contemporary and culturally relevant pt-PT visual text extraction. To ascertain quality, we employ an annotation pipeline combining transcriptions from a frontier LVLM with exhaustive review by native speakers. We observe a sharp performance drop from synthetic to real world samples in most models, and find that, currently, specialized multilingual data is a better driver for pt-PT performance than model size or resolution budget, motivating the release of open pt-PT OCR resources.
arXiv:2512.11895v3 Announce Type: replace Abstract: This work presents an interpretable parametric surrogate model motivated by the need to identify a hydrodynamic model for resolving the trajectory of an object in real-time. The surrogate is formulated as a reduced-order model for a canonical configuration in which a one-degree-of-freedom heaving sphere operates near a vertical wall. High-fidelity CFD simulations are used to generate a parametric dataset of heave-decay responses over varying wall distances (WD) and drop heights (DH). Sparse Identification of Nonlinear Dynamics (SINDy) is then applied to each CFD trajectory to identify a low-order nonlinear ordinary differential equation (ODE) with polynomial terms representing effective hydrostatic restoring and radiation damping, and the harmonic terms representing the wave-induced excitation forces. The SINDy identified coefficients are then used as a prior constraint in a neural operator network (ONet) that learns a smooth mapping from wall distance and drop height to the ODE coefficients, yielding a surrogate capable of predicting dynamics at arbitrary points in the input space without rerunning expensive CFD calculations. The resulting surrogate reproduces CFD heave-decay responses with near-optimal accuracy given the limiting assumptions while being capable of running in real time. The approach provides a practical pathway toward real-time, physics-informed surrogate modelling for launch-and-recovery operations.
arXiv:2607.00524v1 Announce Type: new Abstract: The simulation of locomotion in limbless, deformable organisms remains a challenging problem across computer graphics, soft robotics, and computational modeling. In this work, we present a novel differential-geometric framework for modeling the motion of slender soft bodies, such as snakes. The body is represented as a three-dimensional parametric curve using a Fourier-Chebyshev polynomial basis. Motion is computed by solving an optimization problem that determines the interaction between the curve and its environment by estimating polynomial coefficients. To ensure physically plausible and non-self-intersecting behavior, bending and torsional energy terms are incorporated into the formulation. The resulting curve is then used to drive a surface representation via interpolation, enabling realistic visualization analogous to skinning techniques. We evaluate the proposed approach across a range of complex scenarios and parameter settings to demonstrate its robustness and versatility. Comparative analysis with state-of-the-art methods indicates that our approach achieves improved simulation quality and generates more physically realistic motion.
arXiv:2607.00334v1 Announce Type: new Abstract: Autonomous agents, whether LLM-driven software agents or robotic physical agents, face a common class of failure modes when operating without continuous human oversight: safety violations from unverified actions, behavioral instability from unconstrained loops, and continuity loss from unhandled error states. We develop \system{}, a discrete-time control system that combines five execution gears (\Gobs{}, \Gsug{}, \Gplan{}, \Gexec{}, \Gint{}) with utility-gated dispatch and event-driven fallback. For the single-agent case, we prove monotonic stability, execution safety, eventual stabilization, fallback completeness, and equivalence to a gear-constrained Markov decision process. For multi-agent cyber-physical systems (CPS), we apply the established \smart{} managed-autonomy lifecycle and map runtime evidence into its four governance states (\Stable{}/\Meta{}/\Assisted{}/\Regulated{}). Consensus gating, swarm-level Lyapunov analysis, per-agent gear authority, and rendezvous control provide distributed safety and stability guarantees, including zero collision under the stated assumptions. We evaluate the resulting runtime on a three-agent UR5 robotic assembly cell using fault magnitudes calibrated from the NIST \emph{Degradation Measurement of Robot Arm Position Accuracy} dataset across 10,000 Monte Carlo episodes. It achieves a 99.6\% anomaly detection rate versus 2.1\% for the single-agent baseline, reduces detection latency by $3.5\times$, and supplies a formal physical-workspace safety certificate. The execution gears act as micro-level permissions beneath the \smart{} runtime governance states, separating action control from autonomy governance.
arXiv:2601.11046v2 Announce Type: replace Abstract: Machine learning models for climate and Earth science are becoming increasingly capable, yet model deployment into operational use remains a largely unaddressed challenge: general-purpose model-serving tools, such as MLflow and KServe, assume input data availability at the inference node, while data acquisition, failure handling, and preprocessing are trusted to a separate workflow. We present OpFML: Operational Forecasting with Machine Learning - a configurable pipeline integrating the four steps of operational inference into a single TOML-configured workflow: data consumption, contingency handling, preprocessing, and model inference. By consolidating these steps, OpFML removes the significant boilerplate code required for each new deployment. We demonstrate the pipeline on the operational forecasting of daily fire activity over southern Italy.
arXiv:2606.19100v3 Announce Type: replace Abstract: Large Vision and Language Models (LVLMs) have advanced rapidly, yet European Portuguese (pt-PT) remains systematically underserved by existing open-source multimodal models, which either conflate it with Brazilian Portuguese or severely under-represent it in their training data mixes. We introduce AMALIA-VL, the first open-source instruction-tuned LVLM built natively for pt-PT, pairing a high-resolution vision encoder with dynamic image tiling and a fully open pt-PT-optimized language model via a learned connector. We contribute with a purposefully designed three-stage training process - vision-language alignment, general visual instruction tuning, and preference optimization - together with a pt-PT-centric multimodal data mix combining curated and translated public datasets with novel datasets that address the near-total absence of European Portuguese multimodal resources. Our evaluation shows that AMALIA-VL establishes a strong baseline for open-source pt-PT LVLMs. We will release model weights, training data, and construction pipelines along with machine-translated pt-PT evaluation benchmarks to help democratize pt-PT LVLM development.