Robotics
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Monday 14 September
We present a primitive-informed sampling-based model predictive control (MPC) framework for multi-fingered dexterous manipulation. Sampling-based MPC evaluates candidate control trajectories through forward simulation without requiring gradients through complex contact dynamics. However, directly sampling these trajectories in the high-dimensional joint space of a dexterous hand is inefficient and makes performance strongly dependent on the sampling distribution. Our framework biases sampling using low-dimensional manipulation primitives that encode coordinated finger motions, while simultaneously optimizing joint-level residuals to adapt these motions to the current hand-object configuration. Task-related rollout constraints reject infeasible trajectories during forward simulation, improving the effective use of the sampling budget. We evaluate the approach on a 16 DoF Allegro hand using a synchronized MuJoCo digital twin. Ablations show that both the primitive and residual are necessary for reliable continuous in-hand rotation, that increasing the sampling budget alone does not recover this coordination, and that rollout constraints substantially improve success rate. A primitive extracted for one object size transfers to other sizes and remains effective under model mismatch. The framework further supports grasping, object reorientation, and coordinated arm-hand manipulation, using primitives extracted from both a simulation-trained policy and human hand-motion data.
Commutative Algebra
Generalized divisor topology of commutative rings
Let $R$ be a commutative ring with nonzero identity and let $R^\#$ denote the set of its nonzero nonunits. We extend the divisor topology $D(R)$, previously studied for integral domains, to arbitrary commutative rings and introduce the generalized divisor topology $GD(R)$ on $EC(R^\#)$. Its basic open sets are \[ B_a=\{[b]\in EC(R^\#): b\mid a^n \text{ for some }n\geq 1\}. \] The relation \[ [b]\in B_a \quad\Longleftrightarrow\quad \sqrt{aR}\subseteq\sqrt{bR} \] shows that $GD(R)$ records radical divisibility among principal ideals. We prove that $GD(R)$ is an Alexandrov space and identify its Kolmogorov quotient with the poset of radicals of nonzero proper principal ideals. This description yields characterizations of the $T_0$ and discrete properties and of the equality $GD(R)=D(R)$. We also determine the isolated points of $GD(R)$. Further, we characterize nestedness, compactness, the Lindelöf property, and Noetherianity in terms of the order structure of radicals of principal ideals. In particular, for an integral domain $R$, $GD(R)$ is compact if and only if $R$ is a $G$-domain, while for a UFD the Lindelöf and Noetherian properties are determined by the number of nonassociate prime elements. Finally, we study the interaction of $GD(R)$ with multiplication and describe the behavior of its Kolmogorov quotient under surjective homomorphisms with nil kernel.
Read the full abstract
Differential Geometry
Ricci Solitons, Almost Theta-Yamabe Solitons, and Finite-Order Tensor Symmetries of a Vector Field on Riemannian Manifolds with Rank-One Anisotropic Curvature
We study the iterated action of the Lie derivative on the curvature tensor of a Riemannian manifold with rank-one anisotropic curvature, whose Riemann tensor is expressed via the Kulkarni-Nomizu product as $R = \lambda (\xi^\flat\otimes\xi^\flat)\owedge g$. First, we examine the conditions under which such a manifold admits a Ricci soliton structure and demonstrate that this property implies the almost $\theta$-Yamabe soliton structure. Furthermore, we show that if the associated potential vector field $X$ is a symmetry of the Ricci tensor of a fixed order $k$ (i.e., $\mathcal{L}_X^k \operatorname{Ric} = 0$), the geometric problem reduces to solving a partial differential equation of order $k+1$ along the flow. Finally, under the assumption that $X$ is a conformal vector field ($\mathcal{L}_X g = 2\varphi g$) whose infinitesimal flow preserves the line distribution $\mathcal{D}=\operatorname{Span}\{\xi\}$ (with $[X,\xi]=a\xi$ for $a\in\mathbb{R}$), we prove that several key geometric problems (such as establishing the relation $\mathcal{L}_X^k R = R$, determining the minimal order $k$ for $X$ to be a Lie curvature symmetry, or satisfying $\mathcal{L}_X^{k+1}R = f \mathcal{L}_X^k R$ for a continuous function $f$) are equivalent to a scalar differential problem governed by the operator $D_X = X + 6\varphi + 2a$.
Read the full abstract
Combinatorics
Nullstellensatz degree under Hajós joins and vertex identifications
We study the minimum coefficient degree $N_{k,\F}(G)$ of a Nullstellensatz certificate for Bayer's $k$-coloring equations, where the characteristic of $\F$ does not divide $k$. If $J$ is a \HJ\ join of non-$k$-colorable graphs $G,H$ and $m=\max\{N_{k,\F}(G),N_{k,\F}(H)\}$, then $N_{k,\F}(J)\leq m+k$. When deletion of the selected edge makes each input $k$-colorable, we also have $N_{k,\F}(J)\geq m$; the degree congruence then gives $N_{k,\F}(J)\in\{m,m+k\}$. This partially answers a question of Li, Lowenstein, and Omar. For three-coloring over $\F_2$, we construct an infinite $4$-critical family of exact degree seven, attaining the bound at input degree four. In contrast, every graph constructed from $K_4$ solely by \HJ\ joins has degree $O(\log n)$ and a certificate with polynomially many terms: joins preserve treewidth at most three, and balanced separators yield low-degree certificates. Additional vertex identifications are excluded from this obstruction. We classify all single identifications of the $25$-vertex base graph; exactly $36$ preserve degree seven, producing $24$-vertex $4$-critical graphs of treewidth four. A compressed self-join at adjacent true twins prevents degree loss and gives a repeatable rule adding four vertices per round. The rule does not establish degree amplification or preservation of criticality. Exact witnesses and standalone verification programs accompany the finite results.
Read the full abstract
Artificial Intelligence
GGUF-Metadata Prediction of Single-Sequence llama.cpp Throughput Across Three Systems
We predict single-sequence model throughput from GGUF metadata using roofline-shaped predictors with quantization-specific scale factors fitted on reference models. The scored cohort comprises 318 phase-depth measurements from 53 host-file configurations on two Apple M4 Max systems and an NVIDIA RTX 5080. On host-specific held-out sets of four, five, and two configurations, an active-parameter decode model obtains 13.1%, 14.4%, and 36.1% mean absolute percentage error (MAPE), versus 49.4%, 55.3%, and 51.9% when charging total parameters. Leave-one-host-out coefficients fitted on the other two systems yield 11.6%, 16.8%, and 36.0% test MAPE. A low-bit model ladder changes ordering across runtime stacks. The P2 prefill baseline gives 18.7%, 22.2%, and 108.2% test MAPE. GGUF structure helps on all three systems, but fitted efficiencies are not universal.
Read the full abstract
Artificial Intelligence
Domain Generalization for Smartphone-Based Human Activity Recognition: A Systematic Analysis of Components and Interactions
Smartphone-based Human Activity Recognition (HAR) models often degrade under distribution shifts caused by changes in users, devices, sensor placements, environments, and acquisition protocols. Domain Generalization (DG) addresses this problem by learning from source domains without access to target data. Existing DG methods span training objectives, representation initialization, and architectural modifications, but these components are typically evaluated in isolation despite operating at different stages of the learning pipeline. We present a large-scale controlled benchmark of DG for smartphone-based HAR, comprising more than 410,000 experiments across four model architectures, thirteen training objectives including Empirical Risk Minimization (ERM), five initialization strategies, four architectural configurations, and two shift scenarios: cross-dataset and cross-position. Results show that individual DG components provide limited and highly conditional gains. Alternative objectives rarely outperform ERM consistently, self-supervised initialization helps in specific settings, and architectural modifications, particularly Dynamic Domain Generalization, provide the clearest standalone improvements. Joint configurations, however, frequently outperform their individual components and exhibit complementary and sometimes super-additive interactions, although gains remain model- and shift-dependent. Class-level analysis shows that the strongest configurations mainly improve difficult, shift-sensitive decision boundaries. Finally, oracle checkpoint analysis reveals substantial unrealized performance: source-validation selection recovers only 53% and 26% of the available oracle gain in cross-dataset and cross-position settings, respectively. Overall, effective HAR domain generalization requires jointly designing DG components and robust model-selection strategies.
Read the full abstract
Cosmology and Nongalactic Astrophysics
Field-Level Baryon Acoustic Oscillation Reconstruction of the DESI DR1 Luminous Red Galaxies with Linear Field Transformer (LiFT)
We present the first application of neural field-level baryon-acoustic oscillation (BAO) reconstruction to real spectroscopic survey data. We develop Linear Field Transformer (LiFT), a 3D vision transformer that takes as input the observed galaxy field, its standard reconstruction, and a set of context channels encoding local line of sight, survey coverage, and redshift, and learns to correct standard reconstruction toward the linear density field. We construct a forward-modeling pipeline to produce mock lightcones similar to the DESI Data Release 1 (DR1) luminous red galaxy (LRG) sample, and train LiFT on these. We validate LiFT on held-out simulations, as well as on additional mocks which differ in gravity solver, halo finder, HOD, cosmology, and fiber-assignment history, as well as on mocks analyzed with a distorted distance-redshift relation; ultimately, we find unbiased dilation parameters with consistently tighter constraints than standard reconstruction. Applied to the DESI DR1 LRGs, LiFT improves errors on $\alpha_{\rm iso}$ by 13%, 24%, and 30% and on $\alpha_{\rm AP}$ by 5%, 22%, and 30% relative to the DESI DR1 standard reconstruction analysis in the LRG1, LRG2, and LRG3 bins respectively; meanwhile, our DR1 central values remain consistent with DR1 and DR2. This equates to a factor of 1.2, 1.7 and 2.0 increase in Figure of Merit (or effective survey volume) if one were to only use standard reconstruction. Ultimately, these results establish LiFT as a validated, survey-ready tool for current and upcoming galaxy surveys.
Read the full abstract
Machine Learning
Semantic Fibers and Cross-Gram Interference: A Calculus of Safety Drift in Overcomplete Representations
A deployed language model may refuse a harmful request in English yet comply with its faithful translation, revealing a cross-lingual safety failure that cannot be characterized reliably by output behavior alone. We formalize this phenomenon through an audited equivalence relation and show that, for a declared quotient, representation, metric, feature dictionary, scoring head, threshold, and contrast model, the resulting safety drift admits an exact linear-algebraic characterization. Specifically, the drift is a cross-Gram functional of the within-fiber contrast; its worst admissible value is a support function, while margin invariance is characterized by an annihilator condition. We introduce an intrinsic calibrated exposure measure, governed by the leverage duality $\chi^2=1/\ell-1$, which separates observed drift into three diagnostically distinct regimes: a reader fault removable by recalibration, an exact correction that is too ill-conditioned to be reliable, and a representation-level collision that no readout-only intervention can remove. Thus, identical observed exposure can lead to fundamentally different remediation verdicts. The framework also extends to cone-valued safety heads. An untied order-swap identity provides a diagnostic for the linear control interface; its calibration-state residual predicts a distinct three-control composition error on unseen states and targets, achieving median Spearman correlation $0.964$, compared with $0.269$ for a static cross-Gram baseline. etc.....
Read the full abstract
Computation and Language
One Example Is Enough to Pass Fairness Benchmarks: Rethinking Fairness Evaluation for Aligned LLMs
Warning: This submission studies stereotypes and biases, and contains toxic and offensive examples, used for illustration purposes only. Fairness benchmarks such as BBQ have become the de facto standard for fairness evaluation across major model families. We argue that these benchmarks are too easy to support their role: training Qwen 2.5 7B Base with Group Relative Policy Optimization (GRPO) on a single BBQ example, or placing that example in context as a one-shot demonstration for in-context learning (ICL), lifts mean BBQ accuracy from 79.9% to 92.9% and 99.0%, respectively, closing 80% of the gap to its large-scale RLHF counterpart (96.1%) with GRPO, and surpassing it with ICL. These effects generalize across model families. A cross-conditioning analysis shows the improvement is carried by the reasoning traces generated by the model, and one example suffices to elicit a category-agnostic “missing evidence” reasoning pattern. We argue that BBQ-style multiple-choice abstention benchmarks measure a single structural cue, and a model that solves them does not thereby become fair. We call for evaluation suites that cover a broader spectrum of fairness alignment.
Read the full abstract
Trading and Market Microstructure
Gate Design and Stage-Dependent Incentives in Retail Proprietary-Trading Evaluations: Why Passing Is Not Standalone Evidence of Skill, and Why the Product Fails to Pay Under Measured Trading Constraints
Retail proprietary-trading firms sell a two-stage product: a paid evaluation that must reach a profit target before breaching a trailing drawdown, then a funded account that must survive a minimum window and a consistency rule before a payout. We show the geometry of this contract creates incentives that differ by stage and make passing a poor standalone signal of skill. Under end-of-day trailing the evaluation rewards a fast, lumpy cadence while the funded account punishes it, by a factor of nine in the joint gate. The evaluation is defeatable at zero skill: position sizing alone yields a pass probability near 0.40, against a measured cohort rate of 0.168. Pass probability rises with skill, but a real edge and aggressive sizing move it by nearly the same amount, so a pass rate confounds the two. Under a simplified contract model the seller's margin is bounded by the gap between perceived and actual gate probabilities, the probability analogue of shrouding a price component; the observed design, a permeable marketed evaluation and a hard payout gate, is consistent with that. Across the sector, pass rates are published far more often than payout rates. The same geometry produces negative expected value: within the measured strategy universe, no configuration at observed drifts clears break-even on any account sourced. Break-even lies between a 40.5% and 41.5% win rate at 1:1.5 net of costs, against a driftless 40.0%. The delta-neutral construction firms prohibit is approximately expected-value neutral at the observed payout ceiling. Every account-level result is reported against a zero-edge control.
Read the full abstract
Computation and Language
Dream-RSI: Recursive Self-Improvement through Evolving Worlds
Recursive self-improvement is becoming essential for autonomous AI agents, whose progress depends on discovering high-value solutions across complex domains. Effective exploration drives this process, yet managing and improving exploration strategies remains a major bottleneck. Current systems face a fundamental dilemma: fixed strategies fail to adapt as search spaces scale, while online policy optimization must navigate vast meta-search spaces under delayed, expensive feedback from long-horizon rollouts. We introduce Dream-RSI, a framework for scalable, recursively self-improving exploration. A lightweight orchestration layer makes exploration explicit and programmable while leaving the underlying base agent unchanged. Our key insight is that accumulated discovery history can act as a replay simulator over the realized search space. By dreaming within this simulator built from historical discovery trees, Dream-RSI obtains immediate, low-cost off-policy feedback to evaluate and refine exploration policies without repeated, expensive online evaluation. The improved policy is then redeployed online to drive further discovery, continuously expanding the simulator pool in a self-improving loop. Across 9 tasks in 4 domains, Dream-RSI achieves competitive quality and improves discovery efficiency in several settings.
Read the full abstract
Computation and Language
ModularRSI: Modular and Generalizable Recursive Harness Self-Improvement
Recent work extends recursive self-improvement (RSI) to agent harnesses for long-horizon coding and terminal tasks, enabling agents to improve execution mechanisms from experience. However, generalizable harness RSI remains challenging. First, evolving harnesses on evaluation benchmarks or their subsets makes it difficult to distinguish reusable improvements from benchmark-specific adaptation. Second, single-trajectory updates can conflate systematic harness deficiencies with instance-specific reasoning and solution details, producing modifications that transfer poorly to unseen tasks. Third, localizing recurring behavioral deficiencies within monolithic harnesses is difficult, while whole-harness optimization can entangle unrelated mechanisms and complicate attribution and validation. We propose ModularRSI, a benchmark-disjoint, contrastive, and modular framework for generalizable harness evolution. ModularRSI contrasts successful and failed trajectories for the same task and aggregates evidence across tasks to identify recurring behavioral deficiencies. It decomposes the evolvable harness into five functional modules: Agent Loop, Tool Use, Observation Management, Context Management, and Task Completion Detection. Each module evolves independently within a restricted modification scope, followed by an integration stage that combines the evolved modules into a unified harness and resolves potential conflicts. To support benchmark-disjoint evolution, we curate 2,000 executable evolution tasks from external sources that are disjoint from downstream evaluation benchmarks. Experiments on TB2.0 and SWE-Bench Verified show consistent improvements on unseen in-domain and cross-domain tasks, with the evolved harness also transferring across different foundation models.
Read the full abstract
Computer Vision
RAIN: Region-Aware Inversion Network for Semantic Watermark Extraction
Semantic watermarks for diffusion models embed ownership information into the generative process while preserving perceptual quality, but Gaussian-Shading extraction conventionally requires multi-step diffusion inversion to recover the initial noise. Recent one-step methods show that this cost can be reduced substantially. We study this problem through extended flow matching and conditional regression. The key observation is that, near the high-SNR image endpoint, recovering a useful noise statistic given by the first-step output of the extended flow matching in the high-SNR regime is much simpler than reconstructing the full inverse trajectory, and Gaussian Shading only requires the recovered latent to remain in the correct watermark decision region. Based on this observation, we propose a lightweight, prompt-free extractor that decomposes endpoint recovery into an image-like anchor and a noise-oriented residual, which increases the capability of the model to utilize GPU parallel computation. The resulting method avoids iterative inversion and repeated evaluation of a diffusion-scale U-Net, providing an efficient one-step extraction pipeline with a concise theoretical interpretation. The computational cost of extracting noise is lower than that of both OSI and FARI. The github repo is there: https://github.com/TheLovesOfLadyPurple/RAIN-lightweight-NN-for-one-step-semantic-watermark-extraction
Read the full abstract
Complex Variables
Resonant saddles, tangencies with small spheres, and the characteristic flow
We study resonant Siegel saddles in $\C^2$, with eigenvalue ratio $-p/q$, where $p,q$ are coprime positive integers, through their position relative to small round spheres. An intrinsic invariant $k$, defined through a resonance index that measures approximate first integrals, equals the first resonant obstruction. For $k<\infty$, the maximal contact order with the linear foliation is $m=k(p+q)+1$, and suitably normalized charts give $2k$ tangency circles. We compute their transverse type: resonant terms cancel to first order, and Poincaré–Dulac normalization through degree $2m-1$ makes every contact a saddle, while weaker normalization can produce elliptic arcs and periodic characteristic orbits. For the exact model we describe the whole-sphere flow and the correspondence between its tangency circles and separatrix sectors. In arbitrary charts the characteristic return map near a separatrix circle realizes its holonomy, whose $p$-th iterate has, when $k<\infty$, multiplicity $kp+1$, with $kp$ attracting and $kp$ repelling petals. Classical rigidity of resonant holonomy then recovers the analytic saddle class from $C^1$ orbital equivalence near that circle. We also describe tangency loci for arbitrary nondegenerate germs by the numerical range and the leading orbital form.
Read the full abstract
Sunday 13 September
Artificial Intelligence
Agent Harness Awareness: Turning Verified Telemetry into Operational Control for Tool-Using Agents
Tool descriptions tell an agent what an action can do, but they do not necessarily reveal whether that action is healthy or suitable at the moment of use. Execution harnesses often possess this information through monitors, traces, and recent failures while the model choosing the next action does not. We call this gap operational blindness. We introduce Agent Harness Awareness, a trust-separated design in which the host supplies a small, fresh, verified state snapshot before a consequential action while retaining measurement, enforcement, evaluation, and final dispatch authority. We test its tools-and-network instance with ToolRoute, a controlled isolation benchmark that holds the task and candidate semantics constant and varies the operational information available to the model. In a 216-decision ablation across GPT-5.6 Sol, Claude Sonnet 5, and Gemini 3.7 Flash, verified tool-health state reduced mean observable policy regret by 96.9% against a same-shape neutral control (4,252 ms to 131 ms) and raised single-dispatch completion from 69.4% to 95.8%. In a separate 27-decision end-to-end replication with live local TCP/HTTP execution, OpenTelemetry client spans, and Toxiproxy-injected faults, verified state reduced mean regret from 8,956 ms to 0 ms (zero excess observable operational cost) and completed the live action in 9/9 model-by-fault cases versus 5/9 for the neutral control. These studies establish the tools-and-network instance of Agent Harness Awareness: selected host telemetry materially improves pre-dispatch operational choice. Runtime state, hardware pressure, economics, retry-aware recovery, and heterogeneous selection remain open empirical extensions of the same architecture.
Read the full abstract
Computation and Language
Composable Parametric Demographic Injection for Values Survey Simulation
Large language models (LLMs) are increasingly used as simulated respondents in computational social science, yet accurately reproducing population-specific heterogeneous response distributions remains challenging. Existing methods such as demographic prompting are sensitive to prompt design and increase inference overhead, while parameter-based methods struggle to reuse demographic information efficiently, limiting generalization to unseen combinations. In this regard, we propose PARADE, a composable PARAmetric DEmographic injection framework that represents each demographic attribute value with a reusable and composable LoRA module. Specifically, PARADE separates task-common survey capability from demographic-specific value shifts via a shared Task LoRA and attribute-level Demographic LoRAs, with composition calibration to reduce parameter interference. At inference time, target population models can be constructed on demand by recombining the relevant modules, including demographic combinations not observed during training. Extensive experiments on the World Values Survey and SocioBench across three backbone models show that PARADE achieves strong population-level distribution alignment while maintaining competitive predictive performance, reducing KL divergence by up to 52.8% relative to the strongest baseline.
Read the full abstract
Quantum Physics
Empirical Scaling Laws of Quantum Vision Transformers
Quantum machine learning (QML) is a promising application of quantum computing, yet when and how practical learning models can achieve quantum advantage remains poorly understood. Even a polynomial improvement in the resources needed to attain a given predictive performance would be significant. In this work, we investigate empirical scaling laws of quantum vision transformers (QViTs), which connect QML to the transformer architecture underlying modern large language models. We ask whether QViTs exhibit scaling behavior analogous to classical models and how this behavior changes when qubit number is included alongside dataset size and model parameters. The validation losses in our numerical image-classification experiments are well described by a joint shifted power-law model in qubit number, transformer depth, and training dataset size. Expressing this fit in terms of the classical encoder parameter budget identifies a regime in which increasing the number of qubits can compensate for fewer classical encoder parameters at comparable predicted performance. This finding is consistent with earlier observations of parameter efficiency in quantum learning models and suggests a direction for exploring possible resource savings in QML.
Read the full abstract
Dynamical Systems
Global Regularization of the Spatial Elliptic Restricted Three-body Problem: Quaternionic, Hamiltonian Formulation
Any objects in space may encounter singularity due to the inverse-square law of the gravitational force, hindering the analysis and propagation of particle motion in space. Regularization in celestial mechanics removes such gravitational singularities in equations of motion by transforming the time and coordinates. This paper addresses a class of regularization methods known as global regularization for restricted three-body problems (R3BPs), which simultaneously removes the two singularities (e.g., Earth and Moon in cislunar space). It enables analysis, propagation, and prediction of particle motion throughout the domain of R3BPs, which is in stark contrast to regularizing each singularity individually. In particular, this paper develops a formulation for global regularization of the spatial elliptic R3BP (ER3BP) under external perturbations. The presented formulation leverages mathematical tools from quaternionic (hyper-complex) algebra and Hamiltonian mechanics. This approach enables us to (i) elegantly handle the four-dimensional coordinates arising from the dimension-lifting operation for spatial regularization, (ii) systematically derive the coordinate-momentum transformation while preserving the canonical structure and incorporating perturbing forces, and (iii) provide geometric insights for analyzing the regularity of the transformation, resulting Hamiltonian, and dynamical flow. Numerical examples demonstrate the validity of the mathematical results as well as significant numerical benefits of the regularization formulation in perturbed Earth-Moon ER3BP.
Read the full abstract
Rings and Algebras
Complex slices, inverses, and deformations of a three-dimensional real algebra
We study the real three-dimensional unital algebra with basis $\{1,i,j\}$ and products $i^2=j^2=-1,\quad ij=0,\quad ji=i+j$. Its quadratic and power-associative elements form the union of exactly three complex subalgebras; their nonzero elements are precisely those with a two-sided inverse, which is unique. Every nonzero subalgebra contains the unit, and every element outside these three planes generates the whole algebra. We classify one-sided inverses, including the singular cases, and show that two-sided invertibility does not imply cancellation. The multiplication algebra is $M_3(\mathbb{R})$, so the algebra is absolutely central simple. Its derivations vanish, its automorphism group is trivial, and its commutator Lie algebra is $\mathbb{R}\oplus\mathfrak{aff}(1)$. Its unique antiautomorphism is involutive, and its complexification admits no complex-linear structurable involution. Finally, replacing $ji=i+j$ by $ji=t(i+j)$ gives real algebras $\mathcal{A}_t$ with $\mathcal{A}_s\cong\mathcal{A}_t$ exactly when $s^2=t^2$. We describe the exceptional parameters and thereby exhibit nontrivial deformations of the original algebra.
Read the full abstract
Computational Engineering
A Generalized Source Integral Equation for Homogeneous Penetrable Scatterers
The generalized source integral equation (GSIE)approach is extended to enhance the compressibility of moment matrices for essentially-convex homogeneous penetrable scatterers. In the proposed indirect surface integral equation (SIE) formulation, the scattered field representations use a single distribution of generalized sources that radiate only weakly into the scatterer. A separate distribution of conventional sources is used for representing the interior fields. The wave interactions between surface subdomains via the generalized kernel are of a reduced effective dimensionality. The corresponding exterior integral operator moment matrix blocks exhibit enhanced rank deficiency. For lossy scatterers, this secures the low-rank (LR) compressibility of hierarchical matrix structures for the full system. For scatterers in a much slower background medium, significant savings with very delayed asymptotic scaling of the costs can be achieved. The GSIE is formulated for the transverse magnetic (TM) scattering problem. It is validated for representative examples and its superior compressibility is demonstrated.
Read the full abstract
Computational Engineering
LLM-Based Semantic Modeling and Cooperative Evolutionary Fuzzing for Traffic Violation Scenario Generation
Ensuring the safety of autonomous driving systems (ADS) in a cost-effective and efficient manner remains a critical challenge. Existing law-guided scenario generation approaches are typically limited to a narrow subset of legal rules, resulting in insufficient scenario diversity, and search-based methods often struggle with large and sparse search spaces. To address these limitations, we propose SLaFE (Semantic Law Modeling and Fuzzing based on Cooperative Evolution), a novel traffic violation scenario generation framework designed to systematically evaluate the safety of ADS. SLaFE harnesses the reasoning capabilities of large language models (LLMs) to convert traffic laws into structured scenario constraints. These scenarios are then optimized via a cooperative evolutionary fuzzing algorithm that explores the parameter space to identify boundary cases likely to trigger abnormal ADS behaviors. We evaluate SLaFE on the Apollo platform within the LGSVL simulator using ten real-world traffic regulations. Experimental results show that SLaFE successfully triggered all 10 types of traffic law violations (10/10), outperforming the best existing method, VioHawk (9/10), while others detected no more than 3. Moreover, SLaFE achieved an average triggering time of 5.1 minutes per law type, significantly faster than VioHawk (9.0 minutes) and other baselines. These results highlight SLaFE's effectiveness in discovering diverse and critical law-violating scenarios for ADS testing.
Read the full abstract
Neural and Evolutionary Computing
From Hand-Crafted to LLM-Based Variation Operators in Metaheuristics: A Tutorial
Large language models (LLMs) are increasingly being employed as variation operators in metaheuristics, generating or modifying candidate solutions, heuristics, or programs inside iterative search loops. This shift reframes variation as a model call conditioned on different types of information. We introduce an operator-level framework with two descriptors: (1) the type of prompt-conditioning information at variation time (Numeric, Symbolic, Linguistic), and (2) artifact persistence, identifying what survives the model call (Transient, Amortized, Transfer). The tutorial shows how to classify, build, and select these operators through a worked build template, a method survey, an evidence table, and a cost-aware decision guide.
Read the full abstract
Machine Learning
Energy Vision–Language–Action: A Controlled Multimodal Benchmark for Intent-Conditioned Residential Energy Management
Vision-Language-Action (VLA) models are studied mainly in robotics, where visual observations and language instructions are mapped to physical actions. This paper introduces Energy Vision-Language-Action (EVLA), a controlled multimodal benchmark for intent-conditioned residential energy management. EVLA frames battery scheduling as a multimodal trajectory-prediction problem in which an RGB energy-field representation, a numerical operating state, and a natural-language objective are mapped to a 16-step battery-action trajectory generated by a finite-horizon sampling-based reference generator. Source windows are derived from public residential electrical-load data, while electricity price, battery state of charge, indoor temperature, and time of day are generated benchmark metadata. A hidden operating regime is encoded only through energy-field texture, enabling paired visual changes while the explicit numerical state is fixed. Crossing 439,203 retained base windows with three hidden regimes and five language objectives yields 6,588,045 multimodal instances. An initial study evaluates 36 configurations over three training seeds using fixed subsets of 5,000 training, 500 validation, and 500 test instances. In the MobileNet-family comparison, removing processed language increases trajectory mean-squared error from 0.3856 +/- 0.0039 to 0.8628 +/- 0.0001, whereas removing vision yields 0.3843 +/- 0.0013, comparable to the full model. The results show strong asymmetry in modality use: the processed-language pathway is strongly associated with prediction quality, while the current RGB pathway provides no aggregate error advantage. These results characterize the fixed pilot subset and executed protocol rather than full-benchmark training. EVLA provides a controlled setting for studying how semantic intent and latent context influence residential energy-action prediction.
Read the full abstract
Computation and Language
SignTrace: Describe a Sign, Find the Word
Identifying an unfamiliar sign is difficult when a learner remembers its movement but does not know its meaning or formal feature codes. SignTrace addresses this longstanding reverse-lookup problem through natural-language access to a Chinese sign-language dictionary. The system integrates LLM-based dictionary enrichment, action extraction, dictionary-style rewriting, seven-channel retrieval, and candidate reranking over 6,699 entries. It has been deployed for user trials and has received positive informal feedback. Evaluation on a dictionary-derived benchmark of 500 movement-description queries yields 94.0% Hit@1, 97.4% Hit@9, and a mean reciprocal rank of 0.9540. Reranking increases Hit@1 from 71.8% to 94.0%, while component analyses show the contribution of enriched entry descriptions. Median query-processing time is 13.37 seconds with six concurrent queries. By connecting everyday movement descriptions to documented signs and meanings, SignTrace provides a practical tool for identifying unfamiliar signs. Dictionary-derived wording and prior selection within the benchmark limit generalization to descriptions independently produced by users.
Read the full abstract
Computation and Language
SlideLab: Audience-Centered Scientific Slide Generation and Evaluation
Scientific presentations are more than summaries of research papers. They need to present the work in a coherent sequence, explain the main ideas clearly, and help the audience follow the presentation. We present SlideLab, a training-free multi-agent framework for generating scientific presentations from research papers. SlideLab first plans the presentation narrative, then builds and iteratively refines a shared slide deck using agents for content planning, visual generation, layout refinement, and grounding verification. In a blind human preference study, SlideLab was preferred over both open-source and commercial systems on 77% of papers while using roughly 4 times fewer inference tokens than the strongest open-source baseline. We also introduce ConfArena, an audience-oriented evaluation framework that simulates a conference room and assesses presentations slide by slide. ConfArena matches human system rankings and detects injected presentation problems, including falsified numbers, degraded figures, dropped slides, and shuffled slide order.
Read the full abstract
Computation and Language
Cartograph: Federated Tool Discovery with Operator-Attested Retrieval for AI Agents
The Model Context Protocol (MCP) enables AI agents to discover and call tools, but loading every definition becomes expensive as connected catalogs grow. We present Cartograph, a federated MCP proxy that changes agent-visible tool discovery from $O(n)$ catalog traversal to $O(k)$ progressive disclosure. Cartograph combines three mechanisms: (1) operator-attested capability cards, Ed25519-signed descriptions generated under the deploying operator's control rather than ranked publisher copy; (2) Rift, a three-layer confusable-cluster analysis comprising density clustering, query-margin analysis, and token diagnosis; and (3) two-stage retrieval, which ranks servers before tools. On a 22-server, 374-tool deployment, Cartograph exposes three proxy tools instead of 374 definitions. A 49-query author-constructed benchmark yields R@5 of 0.816, compared with 0.592 for a Jaccard keyword baseline, while a measured top-5 discovery exchange uses 475 tokens rather than 42,450 under the stated full-catalog accounting. Rift identifies 49 confusable clusters, including four HIGH-risk clusters in bootstrap-generated cards. An exploratory comparison of 119 LLM-generated descriptions removes the observed zero-distance cluster but shows that mixing card-generation regimes can reduce R@5. Gateway measurements over ten trials add 5ms mean latency (0.8%) relative to direct stdio MCP calls. Cartograph is complementary to code-execution approaches: it controls which tool descriptions are surfaced and records the provenance of the descriptions used for ranking for each query.
Read the full abstract
Algebraic Geometry
Compact Complex Threefolds Fibred by Two-Dimensional Complex Tori
We construct two families of compact complex threefolds fibred over $\mathbb P^1$ by two-dimensional complex tori. For each positive integer $r$ and each parameter $c$ in a fixed half-plane, the families $Y_{r,c}$ and $Z_{r,c}$ have two multiple fibres of multiplicities $(4,6)$ and $(3,6)$, respectively, and one reduced normal-crossings fibre whose normalization consists of $r$ del Pezzo surfaces of degree six. Their fundamental groups are $\mathbb Z/2$ and $\mathbb Z/3$, respectively, and both have Euler characteristic $2r$. We compute their integral homology; the index-one members are rational homology six-spheres. Every member has algebraic dimension one, lies outside Fujiki's class $\mathcal C$, and has connected automorphism group $\mathbb C^*$. For fixed type and index, two members are biholomorphic exactly when their parameters differ by an element of $(6/r)\mathbb Z$. Each underlying smooth six-manifold therefore supports uncountably many pairwise non-biholomorphic complex structures. The arithmetic input classifies finite monodromy pairs and their integral lifts to a parabolic Jacobi group acting on a rank-four lattice. A quadratic form controls integral lifting and, on the rank-two square-zero locus, computes the monodromy saturation index. We prove that the relevant invariant tensors persist under finite-index restriction, which extends the construction to arbitrary compact base curves. We also construct translation twists after base change, establish a local periodic toroidal quotient theorem in arbitrary dimension, and determine the additive Bockstein spectral sequences of the original threefolds.
Read the full abstract
Numerical Analysis
A Localized Core-Tail Fourier-Laguerre Frame Method with Adaptive Frequency Detection for Unbounded-Domain Problems
We propose a localized core–tail Fourier–Laguerre frame method for approximation and model problems on the real line. The domain is decomposed into a finite core and two semi-infinite tails. Local Fourier extension is used in the core to resolve nonperiodic, oscillatory, and locally nonsmooth structures, while modulated Laguerre frames are used in the tails so that the Laguerre functions approximate only slowly varying decaying envelopes. The local Fourier extension component also provides two data-driven mechanisms: coefficient-energy indicators for internal edge detection and local frequency indicators for selecting tail modulation centers. We derive error estimates that separate the core approximation error, the modulated Laguerre envelope error, the effect of frequency mismatch, and the finite-tail truncation error. The analysis shows that the tail complexity is governed mainly by the residual phase after modulation rather than by the original carrier frequency. Numerical experiments demonstrate high accuracy for oscillatory, multi-frequency, and derivative-discontinuous functions, and a decaying model problem illustrates that the representation can be combined with differential operators through exact interface constraints.
Read the full abstract
Complex Variables
Higher-dimensional singularities of rational inner functions
We study singularities of rational inner functions (RIFs) $\phi$ in three and more variables. We focus on curve and higher-dimensional singularities of $\phi$ in $\mathbb{T}^n$ and we show that, under certain conditions, derivative integrability conditions of partial derivatives of $\phi=\frac{q}{p}$ stay constant along irreducible components of the zero set of $p$, and that the presence of vertical line singularities does not affect global integrability. This answers Question 2 in arXiv:1906.10913. Those results are then generalized to $n$ variables. Finally, we consider slice matrices, which we use to obtain derivative integrability conditions of unidirectional compositions $\phi^N$. We present several new examples, including an RIF with two vertical line singularities and one with vertical surface singularities.
Read the full abstract
Instrumentation and Detectors
A rotation-based Hall-probe magnetic compass
In many physics and space experiments the direction of a magnetic field must be known to one milliradian or better. We have developed a new type of magnetic compass, based on rotation, in which a spinning Hall probe (HP) produces an alternating signal proportional to the component of the magnetic field transverse to the axis of rotation. Aligning the rotation axis so that this signal vanishes gives the direction of the field. The device requires no calibration and is insensitive to electronics drift. A spinning-HP compass was built and used in a recent experiment at Jefferson Lab, where it determined the direction of the 25~G holding field of a polarized $^3$He target to about 1~mrad; the direction of the rotation axis was transferred to the laboratory frame with a laser and a flat mirror attached to the rotor. A second implementation of the rotation approach, a spinning field concentrator (SFC) with a stationary Hall probe, was also built. With 10 seconds of averaging the SFC compass resolves a transverse field of 18~$\mu$G, corresponding to a directional resolution of 70~$\mu$rad in a field of 0.25 G, close to the Earth's field.