arXiv:2604.25698v3 Announce Type: replace Abstract: Tendon-Driven Continuum Robots (TDCRs) pose significant control challenges due to their highly nonlinear, path-dependent dynamics and non-Markovian characteristics. Traditional Jacobian-based controllers often struggle with hysteresis-induced oscillations, while conventional learning-based approaches suffer from poor generalization to out-of-distribution trajectories. This paper proposes a reference-augmented offline learning framework for precise 6-DOF tracking control of TDCRs. By leveraging a differentiable RNN-based dynamics surrogate as a gradient bridge, we optimize a control policy through an augmented reference distribution. This multi-scale augmentation scheme incorporates stochastic bias, harmonic perturbations, and random walks, forcing the policy to internalize diverse tracking error recovery mechanisms without additional hardware interaction. Experimental results on a three-section TDCR platform demonstrate that the proposed policy achieves a 50.9\% reduction in average position error compared to non-augmented baselines and significantly outperforms Jacobian-based methods in both precision and stability across various speeds. For implementation details and source code, please refer to https://github.com/ZiqingZou/ContinuumControl.
Science Journals
arXiv:2607.06312v1 Announce Type: new Abstract: We study adversarially robust algorithms for insertion-deletion (turnstile) streams, where future updates may depend on past algorithm outputs. While recent work achieved a robust $(1+\epsilon)$-approximation for the second moment $F_2$ in polylogarithmic space, achieving high accuracy for other frequency moments remained a major open question; for $p\in[0,2)$, including the fundamental distinct elements problem ($F_0$), only constant-factor approximations were known in sublinear space. We close this gap, showing that $(1+\epsilon)$-approximate robustness can be achieved in polylogarithmic space for all $p\in[0,2]$. Our approach generalizes the estimator-corrector-learner framework to non-Hilbert spaces by dynamically maintaining implicit isometric embeddings into $L_2$ and performing regularized kernel ridge regression over adaptively discovered hard queries, yielding the first insertion-deletion algorithms that approximate: (1) the $p$-th frequency moment $F_p$ up to a $(1+\epsilon)$-factor in poly$(1/\epsilon, \log n)$ space for all $p\in[0,2]$, including the support size $F_0$, (2) metric and information-theoretic quantities, including the Earth Mover Distance (EMD) and $k$-median clustering cost over $[\Delta]^d$ up to an $O(d \log \Delta)$-factor, and the Shannon entropy up to an $\epsilon$-additive error, and (3) non-normed symmetric losses defined by Bernstein functions up to a $(1+\epsilon)$-factor. For the $F_p$ moments, our algorithm is optimal up to poly$(1/\epsilon, \log n)$ factors. Furthermore, we establish a weak equivalence between classical oblivious sketching and adversarial robustness. We prove that for any sub-multiplicative norm, the existence of an efficient classical linear sketch is equivalent to the existence of an efficient robust turnstile algorithm, up to polynomial factors, formalizing $L_1$ embeddability as the fundamental mechanism governing both models.
arXiv:2607.06319v1 Announce Type: new Abstract: Motion understanding is critical for ensuring safety and robustness in autonomous driving systems, driving increasing interest in motion prediction. A key challenge in this domain is the high cost associated with acquiring real-world motion labels. It is therefore ideal if we could transfer motion knowledge from synthetic data to real data. In this context, we explore the potential of synthetic-to-real translation for motion prediction (SRMP). However, the most used naive motion regression methods are notably sensitive to the synthetic-to-real domain shift, resulting in unreliable knowledge translation. To address this, we propose a novel approach integrating a motion knowledge translation framework with two key components: (1) objectness-aware motion prediction, which explicitly models the joint distribution of motion patterns and objectness priors to improve domain-invariant feature learning, and (2) objectness-aided motion enhancement, a motion label refinement mechanism that leverages learned objectness priors to filter motion noise. Furthermore, we present a physically-based pipeline for generating Motion4D, the first synthetic 4D LiDAR dataset tailored for SRMP research, addressing the lack of synthetic motion datasets. Experimental results demonstrate that our approach effectively bridges the domain gaps and yields superior performance on real scenes.
arXiv:2607.06350v1 Announce Type: new Abstract: In confined turbulent rotating convection, the largest vertical velocities are found near the sidewalls in the form of wave-like structures known as wall modes. These structures persist deep into the turbulent regime, bias heat transport, and disrupt bulk flow organisation through radial jets. Controlling or suppressing wall modes is, therefore, essential for accessing bulk dynamics free from wall-induced effects. Here, we combine experiments and direct numerical simulations to investigate wall modes control in cylindrical cells equipped with ring-shaped sidewall barriers. Barriers suppress vertical-velocity maxima near the sidewall and disrupt the characteristic wave-like pattern. Simulations further show that the barriers reduce the wall-mode-induced enhancement of heat transport, shifting it towards values characteristic of laterally periodic domains. The suppression efficiency is governed by the ratio of the barrier width to the wall-mode scale and is enhanced by the addition of a second barrier. In the horizontal plane, radial jet ejections are attenuated, while the time-averaged flow reveals suppression of the boundary zonal flow (BZF), a ring-shaped region of positive azimuthal velocity near the sidewall, provided measurements are taken away from the immediate vicinity of the barriers. In this region, isotherms bend toward the poorly conducting barrier, creating a local misalignment with the isobars and inducing a baroclinic flow adjacent to the barrier faces. This effect weakens with increasing barrier conductivity or smoother geometry. These results demonstrate that sidewall barriers provide a robust route for suppressing wall modes signatures in experimental turbulent rotating convection, while locally inducing secondary baroclinic flows near the barriers. Their use enables access to extreme rotating-convection regimes with reduced sidewall influence.
arXiv:2607.06362v1 Announce Type: new Abstract: In this paper we analyze finite element approximations of ground states of the Gross-Pitaevskii equation in the rapid rotation Thomas-Fermi scaling. In this regime, the healing length and vortex core size are of order $\eps \ll 1$, while the effective confinement potential may degenerate as the angular velocity approaches a critical value. In this setting, we analyze the $\eps$-dependence of the ground states and show that the local flatness of the energy landscape plays a decisive role for numerical resolution. More precisely, we establish mesh size conditions that guarantee the existence of discrete ground states in finite element spaces which are quasi-best approximations of an exact ground state. In particular, we prove that the absolute $H^1$-error behaves asymptotically like $h/\eps^2$. However, to enter this asymptotic regime, the mesh size must satisfy a significantly stronger resolution condition than the natural requirement $h \lesssim \eps$. The additional restriction is governed by the first spectral gap of the Riemannian Hessian of the energy functional at the ground state, which measures the local flatness of the energy surface. With this, our results provide an explanation of the mesh resolution required to capture vortex structures in rapidly rotating Bose-Einstein condensates and highlight the interplay between vortex core size, spectral stability, and discretization accuracy.
arXiv:2606.29248v2 Announce Type: replace Abstract: Vegetable prices in Sri Lanka are highly volatile because the market is largely import-isolated, so supply disruptions quickly drive prices up. This study develops a machine learning framework to forecast such volatility by incorporating supply-chain-aware features and explicitly modelling the country's two cultivation seasons, Maha (October-April) and Yala (May-September). An integrated dataset was constructed by combining retail and farmer-gate prices with origin-aligned weather variables, diesel costs, and exchange rates across 12 vegetable varieties and 14 market centres from 2013 to 2019. A gradient-boosted ensemble model (XGBoost and LightGBM) was trained and optimised using Optuna, and unified and season-specific configurations were compared. Results show that season-specific models improve within-season fit, with the Yala-specific model achieving the highest R2 of 0.9420 (95% CI [0.690, 1.000]), while the unified model delivers the best overall predictive accuracy of 90.84% (95% CI [88.34%, 91.52%]) and an R2 of 0.9281 (95% CI [0.760, 1.000]). Notably, the unified model maintains 85.96% accuracy on a completely unseen 2024 hyperinflationary period without retraining, successfully tracking major price surges. These findings suggest that agricultural price movements in import-constrained markets are meaningfully predictable when models capture supply-chain dynamics, offering practical value for early warning and decision making by farmers, traders, and policymakers. Existing studies on Sri Lankan vegetable prices are confined to Autoregressive Integrated Moving Average (ARIMA) and Generalized Autoregressive Conditional Heteroskedasticity (GARCH) applied to single markets, with no supply-chain features, seasonal segmentation, or cross-regime validation.
arXiv:2607.05315v2 Announce Type: replace Abstract: In this work, we focus on the scenario of a robot-assisted emergency evacuation. We consider two capabilities relevant to such a setting. The first is opening doors ahead of the people being evacuated, so that their path toward an exit stays clear. The second is retrieving rescue equipment and delivering it to the emergency responders carrying out the evacuation. From a systems perspective, this involves several tasks at once. The robot must locate ADA-compliant door buttons and the rescue equipment it needs to retrieve. Additionally, it must remain aware of the people around it and adapt its behavior to them, so that it supports the evacuation rather than getting in the way. We address these demands with a behavior tree at the core of our framework. This structure is chosen for its ability to select high-level tasks based on environmental triggers, and to extend to new situations as they arise. We evaluate the system in 105 trials on the Toyota Human Support Robot, across five hardware and three simulation scenarios. These trials capture the decisions the robot must make in this setting: whether to press a door button, yield to a nearby person, walk through a door someone else is holding, or first retrieve rescue equipment before traversing the door. Overall, the system completes 97 of the 105 trials successfully. These results suggest our framework provides a practical basis for robotic assistance in broader emergency response tasks. Code and video demonstrations are available at https://github.com/AndrewSnowdy/hsr_mm_control.
arXiv:2607.06388v1 Announce Type: new Abstract: Robotic throwing enables fast and efficient object placement beyond the robot's immediate workspace, but reliable throwing in cluttered environments remains underexplored. Existing approaches, such as TossingBot, learn throwing strategies from visual input but assume obstacle-free settings. In this paper, we address the problem of throwing objects into a target basket while avoiding obstacles placed randomly in the scene. We introduce a potential field state representation that compactly encodes both basket attraction and obstacle repulsion on a fixed-size grid, enabling reinforcement learning (RL) policies to generalize across arbitrary numbers and configurations of obstacles. The policy is initialized from kinesthetic demonstrations and optimized in simulation using three state-of-the-art RL algorithms (SAC, DDPG, TD3). Among these, SAC achieves the most consistent performance across scenarios. We compare the potential field representation against explicit state encodings and demonstrate that it achieves higher success rates and better scalability to unseen obstacle configurations. Real-robot experiments with unseen throwable objects confirm robust sim-to-real transfer, achieving up to $90\%$ success in cluttered scenes. These results demonstrate that PFR provides a practical and robust representation for safe and efficient robotic throwing in unstructured environments. A video showcasing our experiments is available at: https://youtu.be/ZZnJf8ua2dE
arXiv:2602.10045v2 Announce Type: replace Abstract: Current instance segmentation models achieve high performance on average predictions, but lack principled uncertainty quantification: their outputs are not calibrated, and there is no guarantee that a predicted mask is close to the ground truth. To address this limitation, we introduce a conformal prediction algorithm to generate adaptive confidence sets for instance segmentation. Given an image and a pixel coordinate query, our algorithm generates a confidence set of instance predictions for that pixel, with a provable guarantee for the probability that at least one of the predictions has high Intersection-Over-Union (IoU) with the true object instance mask. We apply our algorithm to instance segmentation examples in agricultural field delineation, cell segmentation, and vehicle detection. Empirically, we find that our prediction sets vary in size based on query difficulty and attain the target coverage, outperforming baselines (naive best parameter and morphological dilation-based methods). We provide versions of the algorithm with asymptotic and finite sample guarantees. Our work is the first to capture structural uncertainty in instance segmentation by constructing confidence sets of diverse segmentation predictions.
arXiv:2602.10710v2 Announce Type: replace Abstract: With the increasing availability of high-resolution remote sensing and aerial imagery, oriented object detection has become a key capability for geographic information updating, maritime surveillance, and disaster response. However, it remains challenging due to cluttered backgrounds, severe scale variation, and large orientation changes. Existing approaches largely improve performance through multi-scale feature fusion with feature pyramid networks or contextual modeling with attention, but they often lack explicit foreground modeling and do not leverage geometric orientation priors, which limits feature discriminability. To overcome these limitations, we propose FGAA-FPN, a Foreground-Guided Angle-Aware Feature Pyramid Network for oriented object detection. FGAA-FPN is built on a hierarchical functional decomposition that accounts for the distinct spatial resolution and semantic abstraction across pyramid levels, thereby strengthening multi-scale representations. Concretely, a Foreground-Guided Feature Modulation module learns foreground saliency under weak supervision to enhance object regions and suppress background interference in low-level features. In parallel, an Angle-Aware Multi-Head Attention module encodes relative orientation relationships to guide global interactions among high-level semantic features. Extensive experiments on DOTA v1.0 and DOTA v1.5 demonstrate that FGAA-FPN achieves state-of-the-art results, reaching 75.5% and 68.3% mAP, respectively.
arXiv:2607.05985v1 Announce Type: new Abstract: This paper presents a black-box evaluation framework to systematically assess the ability of Large Language Models (LLMs) to generate Design Structure Matrices (DSMs) from structured technical documentation. Motivated by the closed-source nature of current Auto-DSM pipelines, the framework introduces a reproducible methodology that benchmarks generated DSMs (GEN-DSMs) against manually validated ground-truth matrices (GT-DSMs). The evaluation integrates both single-run and multi-run perspectives, combining structural metrics (Completeness, Correctness, Coupling Density), classification metrics (Selective Accuracy, Abstention Coverage), and stability measures (Entropy, Fleiss' $\kappa$). To synthesize these aspects, a Composite Quality Score (Q) is proposed. Controlled experiments are conducted on two datasets: a fictive abstract system and a real-world refrigerator decomposition, covering variations in phrasing, parameter-dataset alignment, and system complexity. Results show that LLMs can produce structurally plausible DSMs and achieve high reproducibility under well-structured inputs, but remain sensitive to ambiguity, inconsistent dependency definitions, and prompt formulation. The findings highlight systematic sources of hallucination and abstention failure, demonstrating both the potential and current limitations of LLM-driven DSM automation. The proposed framework provides a transparent benchmark for auditing Auto-DSM pipelines and establishes foundations for integrating LLM-based decomposition methods into model-based systems engineering (MBSE) workflows.
arXiv:2603.21664v3 Announce Type: replace Abstract: While Omni-modal Large Language Models have made strides in joint sensory processing, they fundamentally struggle with a cornerstone of human interaction: deciphering complex, multi-person conversational dynamics to accurately answer ``Who said what and when.'' Current models suffer from an ``illusion of competence'' -- they exploit visual biases in conventional benchmarks to bypass genuine cross-modal alignment, while relying on sparse, low-frame-rate visual sampling that destroys crucial high-frequency dynamics like lip movements. To shatter this illusion, we introduce Visual-Registered Speaker Diarization and Recognition (VR-SDR) and the HumanOmni-Speaker Benchmark. By strictly eliminating visual shortcuts, this rigorous paradigm demands true end-to-end spatio-temporal identity binding using only natural language queries. To overcome the underlying architectural perception gap, we propose HumanOmni-Speaker, powered by a Visual Delta Encoder. By sampling raw video at 25 fps and explicitly compressing inter-frame motion residuals into just 6 tokens per frame, it captures fine-grained visemes and speaker trajectories without triggering a catastrophic token explosion. Ultimately, HumanOmni-Speaker demonstrates strong multimodal synergy, natively enabling end-to-end lip-reading and high-precision spatial localization without intrusive cropping, and achieving superior performance across a wide spectrum of speaker-centric tasks.
arXiv:2603.22578v3 Announce Type: replace Abstract: Trapped ions are one of the most advanced platforms for quantum technologies, with applications ranging from quantum computing to precision timekeeping. A crucial step towards more compact and scalable systems involves integrating photonic integrated circuits (PICs) into surface ion traps to enable on-chip light delivery and optical addressing of individual ions. Currently, most implementations rely solely on transverse-electric (TE) mode grating couplers, where the emitted light is polarized in the plane of the chip. In this work, we design, fabricate and characterize silicon nitride (Si\(_3\)N\(_4\)) PIC components, including incoupling structures, splitters, and grating couplers that support both TE and transverse-magnetic (TM) modes with comparable optical losses. We benchmark the PIC at 760\,nm, which is a typical wavelength for Yb$^{+}$-applications. The fabricated grating couplers enable the outcoupling of collimated free-space beams for both polarizations, exhibiting distinct emission angles. This dual-polarization capability gives more flexibility in polarization control and expands the accessible optical design space for trapped-ion quantum technologies.
arXiv:2607.06459v1 Announce Type: new Abstract: This paper studies input-to-state stability (ISS) certification for data-driven Koopman learning control of unknown discrete-time nonlinear repetitive systems over finite trial horizons. Rather than proposing a new learning law, we certify when a fixed Koopman-assisted constrained update yields practical stability of the selected tracking error along the trial axis. Prediction accuracy alone is insufficient for this purpose: the selected finite-horizon input-output channel must have a positive margin, and the unreachable component of the requested output increment must be accounted for through a projection residual. Thus, a Koopman predictor with small held-out prediction residuals may still fail the learning-stability certificate if its selected channel is weak. We formulate the selected stacked tracking error as the state of a discrete-time learning-axis system and treat Koopman residuals, reset mismatch, channel uncertainty, projection residuals, deployment shifts, and numerical tolerances as ISS inputs. The deterministic result gives a practical ISS estimate from the initial learning error to an explicit ultimate band. A finite-sample implementation constructs an episode-level residual bound under a fixed controller and combines it with reported channel, projection, shift, and numerical margins. Numerical checks on nonlinear repetitive systems support the predicted residual-to-band scaling, weak-channel rejection, projection closure, and ultimate-band coverage.
arXiv:2607.06464v1 Announce Type: new Abstract: Automated progress monitoring on construction sites is an active area of research and development. Robot and human-carried mapping systems have been developed to build 3D maps of building and infrastructure projects. While LiDAR-based mapping systems achieve high accuracy, the cost of LiDAR can be prohibitive. Consumer-grade cameras with wide field of view ("360 cameras") combined with embedded inertial measurement units (IMUs) provide a cost-effective alternative. To support change detection and progress monitoring, highly accurate visual Simultaneous Localization and Mapping (SLAM) and floor plan-referenced localization systems are required. In this paper we present a high-quality dataset collected at an active construction site, which captures realistic challenges such as variable lighting conditions, moving workers, fast motions, and repetitive structures. The dataset offers thirty visual-inertial sequences recorded across seven floors over an eight-month period of the construction project. Ground truth trajectories were collected using a high quality LiDAR-inertial SLAM system rigidly attached to the 360 camera. Additionally, we report the results of an open research challenge evaluating the best visual SLAM and localization systems from around the world. The Challenge attracted substantially higher participation in SLAM, with 62 teams compared to 22 in floor-plan-referenced localization, reflecting the broader maturity of SLAM methods. The higher errors in localization further highlight the difficulty of this task in construction and point to the need for continued research, which this dataset is intended to support. The dataset and the benchmark are publicly available at: https://hilti-trimble-challenge.com/dataset-2026.
arXiv:2607.05753v1 Announce Type: new Abstract: Compact representations of rarefied flows must preserve kinetic observables, not only smooth macroscopic fields. We introduce Gaussian kinetic representations for discrete velocity method (DVM)-Shakhov solutions of normal shocks and a lid-driven cavity. A positive log-density phase-space model reconstructs shock velocity distribution functions (VDFs) and their moments, while a moment-field model compresses wall-bounded cavity structure. Log-density training recovers heat flux, stress, and third- and fourth-order shock moments without explicit moment supervision; the cavity representation gives a compact continuous wall-transport map.
arXiv:2607.05769v1 Announce Type: new Abstract: We propose a novel pipeline, Legato 2, for extracting symbolic notation and semantic knowledge from images of sheet music. Legato 2 features the first large-scale neural model for optical music recognition (OMR) to operate sequentially on a system-by-system basis, following the horizontal lines of notation as they are read on the page, rather than treating the page as an undifferentiated image, enabling better scaling to arbitrarily long inputs. It is also the first OMR model capable of generating symbolic transcriptions that include embedded textual content, such as titles and annotations. The pipeline combines system-level segmentation with an autoregressive vision-LM to capture both local notation details and score structure. Across multiple datasets, Legato 2 consistently outperforms prior state of the art. We also show that symbolic transcriptions complement visual inputs for frontier language models, improving their interpretation of dense musical documents. Legato 2 establishes new state-of-the-art performance in both OMR and downstream sheet music understanding.
arXiv:2607.06115v1 Announce Type: new Abstract: In next-generation networks, communication systems will no longer be limited to data transmission and will be expected to acquire awareness of the surrounding environment. This leads to the concept of integrated sensing and communication (ISAC), where the same wireless infrastructure is used for both communication and environmental sensing. Thus, ISAC enables the system to transmit information efficiently and observe and interpret channel variations and user behavior. Motivated by this capability, this work focuses on detecting an active attacker in an urban environment scenario, where the attacker intentionally manipulates beamforming directions to increase interference and mislead the transmitter into allocating the main lobe of beam toward itself instead of legitimate users. We apply game-theoretic approaches to model the interaction between legitimate users and the attacker, and integrate the resulting utility-based formulation into a reinforcement learning (RL) framework. Simulation results demonstrate that the proposed method effectively addresses security challenges in dynamic 6G ISAC systems.
arXiv:2604.05543v2 Announce Type: replace Abstract: Multivariate time series forecasting often struggles to capture long-range dependencies due to fixed lookback windows. Retrieval-augmented forecasting addresses this by retrieving historical segments from memory, but existing approaches rely on a channel-agnostic strategy that applies the same references to all variables. This neglects inter-variable heterogeneity, where different channels exhibit distinct periodicities and spectral profiles. We propose CRAFT (Channel-wise retrieval-augmented forecasting), a novel framework that performs retrieval independently for each channel. To ensure efficiency, CRAFT adopts a two-stage pipeline: a sparse relation graph constructed in the time domain prunes irrelevant candidates, and spectral similarity in the frequency domain ranks references, emphasizing dominant periodic components while suppressing noise. Experiments on seven public benchmarks demonstrate that CRAFT outperforms state-of-the-art forecasting baselines, achieving superior accuracy with practical inference efficiency.
arXiv:2604.25301v2 Announce Type: replace Abstract: Job-scheduling games have traditionally assumed fixed processing times. However, in many realistic environments, ranging from cyber-security response to high-frequency trading, a task's duration depends on its starting time. We study job-scheduling games with time-dependent processing times, where job lengths are linear functions of their start times, exhibiting either positive deterioration (increasing length) or negative deterioration (decreasing length). We analyze these games under various coordination mechanisms and priority policies. By introducing the concept of delay-averse agents, we provide a unifying framework to characterize equilibrium existence. For delay-averse jobs, we show that stability is maintained and pure Nash equilibria (NE) can be computed efficiently. In contrast, for non-delay-averse jobs, we demonstrate that a NE may not exist, and prove that deciding its existence is NP-complete, even on identical machines - a fundamental departure from classical coordination mechanisms. Regarding equilibrium inefficiency, we show that the Price of Anarchy (PoA) can be significantly higher than in environments with fixed processing times. To mitigate this, we propose and analyze three coordination mechanisms: SBPT (Shortest Basic Processing Time), which reduces the PoA in games with positive deterioration to a constant, and SDR (Smallest Deterioration Rate) and LBDR (Largest Basic-Deterioration Ratio) for negative deterioration, which achieve tight constant PoA bounds of $2$ and $\max\{\frac{e}{e-1}, 2-\frac{1}{m}\}$, respectively. Our results bridge the gap between centralized time-dependent scheduling and decentralized game-theoretic analysis.
arXiv:2607.06531v1 Announce Type: new Abstract: - Objective: Multimodal deep learning models in oncology are currently limited by monolithic designs that rigidly couple data ingestion, clinical routing, and artificial intelligence (AI) inference. To address this inflexibility, we propose the Large Cancer Assistant (LCA), a model-agnostic, post-hoc orchestration framework designed for scalable clinical decision support. - Methods: The LCA is mathematically formalized as a 7-tuple architecture grounded in the principle of Algorithmic Impermeability, ensuring the orchestration logic remains strictly independent of underlying black-box AI models. We introduce the Entry Theory, leveraging Geometric Deep Learning (GDL) to standardize multimodal patient data along distinct structural and medical axes. The system dynamically orchestrates data via a Cancer Switching Module and intentionally isolates the core AI execution from volatile hospital IT infrastructures by outputting a Standardized Intermediate Payload (SIP). - Results: A Proof of Concept (PoC) validated the orchestration logic across four technical scenarios. The framework executed a nominal flow with negligible orchestration overhead. It empirically demonstrated algorithmic impermeability by maintaining an invariant routing projection during AI model swaps, and it validated strict failure-safety by achieving a 100\% recall rate in generating targeted Supplementary Data Requests (SDR) under injected data anomalies. Multi-protocol execution capability was also successfully verified. - Conclusion: By structurally decoupling multimodal ingestion from feature inference, the LCA provides a highly adaptable and modular orchestration foundation. The SIP establishes a clear architectural boundary, natively setting the stage for downstream Electronic Medical Record (EMR) interoperability as an independent future paradigm.
arXiv:2604.11290v2 Announce Type: replace Abstract: Synthesizing supervised finetuning (SFT) data from language models (LMs) to teach smaller models multilingual tasks has become increasingly common. However, teacher model selection is often ad hoc, typically defaulting to the largest available option, even though such models may have significant capability gaps in non-English languages. This practice can result in poor-quality synthetic data and suboptimal student downstream performance. In this work, we systematically characterize what makes an effective multilingual teacher. We measure intrinsic measures of data quality with extrinsic student model performance in a metric we call Polyglot Score; evaluating 10 LMs across 6 typologically diverse languages, generating over 1.4M SFT examples and training 240 student models. Among the models tested, Gemma 3 27B and Aya Expanse 32B emerge as consistently effective teachers across different student base model families. Further analyses reveal that model scale alone does not significantly predict teacher effectiveness; instead, data qualities such as prompt diversity, length, and response fluency capture over 93.3% of variance in intrinsic data quality and predict student performance. Finally, we provide practical recommendations, including matching the model families of teacher-student pairs and translating from or responding to existing prompts, which can yield improvements for less-resourced languages. We hope that our work advances data-centric research in multilingual synthetic data and LM development.
arXiv:2605.13362v3 Announce Type: replace Abstract: Computational social choice and algorithmic decision theory offer rich aggregation theory but no end-to-end process for egalitarian self-governance: aggregation, deliberation, amendment, and consensus are each considered in isolation, with key metric-space aggregators being NP-hard. Here, we propose \emph{constitutional governance in metric spaces}, integrating these stages into a protocol for constitutional governance. A community's \emph{legal corpus} comprises its \emph{laws} together with a \emph{constitution}, each being a point in a metric space, with the constitution specifying the supermajority threshold required to amend it. Members vote to amend the legal corpus by proposing their ideal points, followed by rounds of submitting \emph{public proposals} carrying \emph{supermajority public support}; a polynomial-time \emph{aggregation rule} scores each proposal, and a supported proposal whose score is positive and maximal for two rounds is adopted; if none is found the status quo is retained. Public proposals can be sourced from deliberation, vote aggregation, or AI mediation. With Constitutional Consensus, a community can run the constitutional governance protocol on members' personal computing devices (e.g., smartphones), achieving digital sovereignty. By drawing on metric-space aggregation, reality-aware social choice, supermajority amendment, constitutional consensus, deliberative coalition formation, and AI mediation, we provide a comprehensive framework for the constitutional governance of digital communities and organisations.
arXiv:2607.06118v1 Announce Type: new Abstract: As web agents increasingly demonstrate capabilities in automated task execution, the development of robust evaluation frameworks for assessing their navigation and task completion performance has emerged as a critical research priority. However, existing benchmarks exhibit fundamental limitations. First, they suffer from insufficient scale and limited domain diversity, constraining comprehensive evaluation of cross-domain generalization. Second, prevailing LLM-as-Judge evaluation methodologies inadequately capture fine-grained interaction semantics, particularly regarding precise query formulation and filtering operations. Third, current benchmarks predominantly emphasize navigation success metrics while neglecting critical requirements for real-world deployment scenarios. To address these limitations, we introduce WebRetriever, a large-scale benchmark encompassing 800 websites and 1,550 tasks across diverse domains, including consumer, professional, and enterprise sectors, with comprehensive coverage of user intent patterns. We propose NavEval (Navigation Evaluation), a novel LLM-as-Judge framework that leverages rich interaction context beyond visual screenshots, achieving state-of-the-art alignment with human judgment across multiple evaluation datasets. Furthermore, we establish three complementary evaluation protocols that collectively provide holistic assessment of web agent capabilities: navigation proficiency, knowledge-assisted interaction, and end-to-end task completion with information extraction. Extensive experimental analysis reveals substantial performance disparities across evaluation protocols, demonstrating that navigation success alone is an insufficient predictor of real-world application effectiveness. WebRetriever delivers fine-grained diagnostic insights into agent capabilities and establishes a rigorous foundation for advancing web agent research and development.
arXiv:2607.06153v1 Announce Type: new Abstract: A proportional wealth tax acts as a uniform gravitational field on the wealth distribution: it shifts the drift of the Fokker-Planck equation without altering the diffusion, preserving the Gini coefficient at all finite times. The same drift-shift symmetry that makes the tax non-distortionary also makes it non-redistributive through the market channel. Redistribution requires breaking this symmetry. A progressive tax (confining potential) replaces the Pareto steady state with a thinner-tailed distribution whose Gini is a closed-form function of the progressivity parameter; source-sink terms (tax-funded transfers) reshape the density directly. We formulate optimal redistribution as a control problem for the Fokker-Planck equation, penalising intervention costs including migration, evasion, and portfolio distortion. In general equilibrium the tax design feeds back through aggregate capital and the production function, yielding a self-consistent McKean-Vlasov equation with diminishing returns to progressivity. The spectral gap of the Fokker-Planck operator determines convergence speed: progressive taxes redistribute within policy-relevant timescales, whereas proportional taxes rely on slow demographic turnover.