Astrophysics of Galaxies
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Monday 7 September
We investigate the geometry of extreme Kerr-Newman black holes and its role in selecting distinct black hole states. Motivated by Smarr's question of what physical information is encoded in event-horizon symmetries, we identify, besides the Smarr extreme family, a special family in which the mass, charge, angular momentum, and irreducible mass are constrained by a single irrational number. In the Christodoulou diagram, this family is selected by a discrete symmetry of appropriately scaled charge and angular-momentum variables. The local differential geometry of the event horizon provides an additional selection principle, which, up to the discrete symmetries (Q \rightarrow\pm Q) and (J \rightarrow\pm J), identifies a unique extreme Kerr-Newman black hole exhibiting the highest degree of local spherical symmetry compatible with the rotating Kerr-Newman geometry. Remarkably, for this distinct extreme configuration, all relevant physical and geometrical quantities, including the energy, electric charge, angular momentum, and irreducible mass, are solely related through the golden ratio. We also examine reversible and irreversible transformations and the associated extractable energy. Intrinsic horizon geometry can therefore constrain the macroscopic parameters and single out distinct extreme Kerr-Newman configurations.
Computation and Language
Legal text classification in Korean sexual offense cases: from traditional machine learning to large language models with XAI insights
The advancement of natural language processing (NLP) has expanded AI-based text classification in the legal domain. However, accurately classifying legal documents remains challenging due to the complexity of legal texts and subtle differences between legal categories. This study evaluates legal text classification models ranging from traditional machine learning techniques to large language models (LLMs) using ten categories of Korean sexual offense precedents. The results show that fine-tuning small-scale models such as KLUE-BERT on legal data outperforms general-purpose models such as GPT-3.5 and GPT-4.0, as well as traditional machine learning models. KLUE-BERT achieved the highest accuracy of 99.3%, indicating that domain adaptation and fine-tuning can be more important than model size for legal document classification. We further employ explainable AI (XAI) techniques to analyze model predictions and misclassification cases. XAI analysis identifies linguistic features influencing model decisions and limitations in capturing subtle textual cues. Using KICS data, which closely resembles real-world legal case records, we further evaluate the model's generalization capabilities and find that it struggles to interpret implicit contextual cues. These findings highlight the importance of both performance and interpretability in legal AI and demonstrate how XAI can improve transparency in legal text classification. AI-assisted tools can support legal professionals in tasks including document classification, legal information retrieval, and case assessment.
Read the full abstract
Combinatorics
A Counterexample to Wormald's Conjecture
We give a counterexample to Wormald's conjecture. A 16-vertex cubic graph, the disjoint union of $K_{3,3}$ with a 10-vertex bridged graph built from two subdivided copies of $K_4$, has no partition of its edges into two isomorphic spanning linear forests. The obstruction is parity: any such partition would force every monochromatic component to have even order while requiring a five-vertex side of a bridge to be partitioned into such components. Adjoining copies of $K_4$ gives counterexamples in every order $16+4t$. The counterexample is disconnected and leaves the connected case unresolved. The full family is verified in Lean without custom axioms.
Read the full abstract
Human-Computer Interaction
Critsly and StudioCrit: An Artefact-Aware AI Critique Workspace and Simulation-Based Readiness Study for Design Education
Critique in design education depends on interpreting work in progress, articulating intentions and translating feedback into revisions. This technical report presents Critsly, an artefact-aware AI critique workspace, and StudioCrit, its architecture-studio research mode. Critsly combines a visual board, design-intention fields, guided reflection, perspective-based critique and action planning. StudioCrit adds studio/class organisation, role-based access, cognitive and architectural classification, educator analytics and exportable evidence. The report consolidates implementation and simulation evidence recorded in a research project submitted in July 2026. Three simulated studio scenarios yielded 109 classified evidence rows, including 85 assigned to higher-order Bloom categories. A separate rehearsal using 50 disposable learner accounts yielded 56 evidence rows, including 46 assigned to higher-order categories. A subsequent hardening rehearsal recorded 50 completed sessions, 50 successful board pulls and 50 denials of student access to analytics. These are software and synthetic-trace observations, not measurements of learning gains or human cognitive performance. Automated classifications remain provisional, and the source report does not establish classifier accuracy or inter-rater reliability. The contribution is an implemented critique-to-evidence workflow and a bounded account of its readiness for further controlled evaluation.
Read the full abstract
Artificial Intelligence
Scientific Agents: Evaluating Profession-Specific System Prompts on Scientific Tasks
Detailed profession-specific system prompts raise token use and estimated cost per response without a consistent accuracy gain. We evaluate Scientific Agents, an open-source corpus of 503 profession-specific AGENTS.md profiles, with Gemini 3.8 Flash via OpenRouter in the Pi agent harness. We compare matched profiles with four controls: a minimal baseline ("You are a helpful assistant"), the profile's opening role sentence, a generic scientific rigor guide, and a profile from an unrelated domain. Across nine text-based science benchmarks (4,531 sampled questions, 100 matched profiles), 4,488 items completed all five conditions after API-error retries, scored with automated, rule-based grading. The average profile-baseline accuracy difference is -0.6 percentage points (95% bootstrap interval [-1.5, +0.2] across fixed tasks), and no benchmark shows a statistically clear improvement. Matched profiles produced 1.5-2.3 times as many output tokens and cost 2.2-4.5 times more per successful call. On 60 tool-using BioMysteryBench bioinformatics problems (three runs each for baseline and profile), mean solve rates were 46.7% with the profile and 56.7% at baseline, a difference of -10.0 percentage points (95% interval [-16.7, -3.3]) driven by more frequent token- and time-limit stops under the profile. Longer prompts had one unexpected operational advantage: on SuperGPQA, frequent provider API drops left the short baseline with a correct first-pass answer on only 54.0% of items, against 71.6% with the profile. Generic and mismatched prompts were about as reliable, so this gain comes from prompt length or formatting rather than domain expertise. For the tested model and tasks, loading full profession profiles by default does not improve accuracy and costs considerably more; whether selective retrieval of profile sections or open-ended scientific tasks would change this remains to be tested.
Read the full abstract
Algebraic Geometry
The GIT of 4x3x3 tensors (determinantal cubic surfaces)
A general $4\times3\times3$ tensor defines three determinantal varieties in $\mathbb{P}^3$ and $\mathbb{P}^2$ and $\mathbb{P}^2$ where it drops rank: a cubic surface and two length 6 subschemes. We study the Geometric Invariant Theory (GIT) quotient of $\mathbb{P}(\mathbb{C}^4\otimes\mathbb{C}^3\otimes\mathbb{C}^3)$ by $\operatorname{SL}(4)\times\operatorname{SL}(3)\times\operatorname{SL}(3)$, a moduli space of linear determinantal representations of cubic surfaces. We give a geometric description of the (semi)stable locus and some of the geometry of this GIT quotient. This GIT quotient admits a surjective rational map to the GIT moduli space of cubic surfaces, undefined at only one point.
Read the full abstract
Algebraic Geometry
Bulk-deformed enumerative mirror symmetry for toric Fano surfaces
We prove a version of genus-0 mirror symmetry for toric Fano surfaces at the level of big quantum cohomology. The mirror Landau-Ginzburg model $(\check{X}, W_{\infty})$ of such a surface $X$ can be defined in terms of counts of Maslov index two tropical or holomorphic discs with bulk insertions. While these mirror LG models are not well-defined, the Jacobian ring $\operatorname{\mathsf{Jac}}(W_{\infty})$ of the superpotential $W_{\infty}$ is well-defined. We show that $\operatorname{\mathsf{Jac}}(W_{\infty})$ is isomorphic over the Novikov base to the big quantum cohomology of $X$ using tropical methods, providing a purely algebraic analogue of the cobordism argument used in works of Fukaya-Oh-Ohta-Ono.
Read the full abstract
Combinatorics
On the number of regular simplices in odd dimensions
Let $S^k_d(n)$ denote the maximum number of regular $(k-1)$-simplices spanned by $n$ points in $\mathbb{R}^d$. For all fixed $r\geq k\geq3$, the exact value of $S^k_{2r}(n)$ was recently determined by Dumitrescu and the authors for all sufficiently large $n$ when $k=3$, and conditionally on an optimization problem when $k\geq4$. In this paper we establish a sharp estimate in odd dimensions by proving that \[S^k_{2r+1}(n)=\binom{r}{k}\left(\frac{n}{r}\right)^k+\Theta(n^{k-2/3})\] for all fixed $r\geq k\geq3$. This can be viewed as a generalization of the result by Erdős and Pach on unit distances in odd dimensions. We also establish a structural characterization of nearly extremal point sets. A notable difference from the even-dimensional case is that nearly extremal point sets in odd dimensions admit two distinct configurations. The proof leverages techniques from hypergraph Turán theory and linear algebra, with additional geometric arguments.
Read the full abstract
Computation and Language
MoM: Memory of Memory
For a long-horizon LLM agent, the memory question is not what was once recorded but what currently holds. Most designs answer it only indirectly: every interaction is stored, and the present is reconstructed at query time by retrieving and reconciling records, so stale values re-enter and the same conflicts are re-litigated. Committing the current value at write time avoids this, but existing write-time (CRUD) memories overwrite, so a wrong update is unrecoverable and prior state is lost. We take the missing combination—commit on arrival while retaining what is displaced—and formalize it as Memory of Memory (MoM): memory tracks not only content but the provenance, status, and history of its own entries. We instantiate MoM as Provenant Memory (P-Mem), a typed provenance graph whose active frontier exposes one current value per resolved key while displaced values are retained as provenance; typed operations decide whether a new observation supports, supersedes, contests, rejects, revokes, or resolves an existing value. P-Mem's decisive gain is validity rather than accuracy: its turn-level read matches the strongest retrieval memory in accuracy at $\sim$4$\times$ fewer read tokens—a retrieval-granularity effect—while graph-guided turn pruning cuts the knowledge-update stale-answer rate (19.4%$\rightarrow$10.9%); on revision chains it stays at 100% where query-time reading collapses to 25%, and, because displaced values are retained rather than overwritten, it recovers committed errors a CRUD memory cannot (100% vs.\ 0%).
Read the full abstract
Computation and Language
LatentPort: Beyond KV Cache - Cross-Model Transfer of Recurrent Memory in Hybrid Language Models: A 4B-to-9B Hybrid-State Handoff Without Target Prefix Replay
Can one language model hand its live memory to another without the receiver rereading the context? We demonstrate useful persistent hybrid-state transfer across one architecture-matched Qwen3.5 4B-to-9B sibling pair. To our knowledge, this is the first demonstrated cross-model handoff of persistent recurrent inference state between differently sized hybrid language models without target prefix replay. Translated attention KV alone leaves a large gap; adding the Gated DeltaNet (GDN) persistent-state package lowers teacher-forced negative log-likelihood (NLL), the average next-token log-loss, by 0.747 nats/token (95% paired document bootstrap CI [0.6921, 0.8047]), improving all 64 PG19 documents. Direct recurrent and convolution reuse outperforms the tested learned GDN maps, consistent with partial functional compatibility of persistent-state coordinates. A fresh component factorial selects translated KV with direct recurrent and convolution state. An additional 434,176-parameter correction improves that base on 64 fresh web documents: continuation loss is 0.076 nats/token above native 9B (excess NLL), Jensen-Shannon (JS) divergence is 0.022, and native context recovery (NCR) is 0.918. Corrected 9B significantly beats continued 4B inference while processing zero historical prefix tokens. Evidence covers one direction, one geometry-matched Base-model pair, and 4K teacher-forced continuation; the near-native gate failed, the 16K branch was not run, and free-generation equivalence and a general state interface remain unproven.
Read the full abstract
Neurons and Cognition
Exploring the robustness of permutation entropy analysis to differentiate between closed-eyes and open-eyes resting states
Electroencephalography (EEG) is a noninvasive technology that is widely used to monitor brain states, and many efforts are focused on developing reliable and efficient data analysis methods for EEG recordings. Here, we apply ordinal analysis to the EEG recording of the resting state of 109 healthy subjects measured in two different conditions: with eyes closed (EC state) or eyes open (EO state). We study the robustness of the temporal permutation entropy ($PE$) and the spatial permutation entropy ($SPE$) with respect to the presence of blinking artifacts in the EO recordings, the duration of the recordings, and the number of EEG channels analyzed. We perform a paired statistical test to assess whether these quantities differ significantly in the EO and EC recordings of the same subject. We find that $PE$ and $SPE$ perform surprisingly well, as they detect significant differences when calculated from the raw signals, even within a short time interval ($PE$) or from a reduced number of electrodes ($SPE$).
Read the full abstract
Multimedia
Hi-Singers: A Comprehensive High-Quality Dataset for Expressive Audio-Driven Singing Head Synthesis
State-of-the-art models for audio-driven digital human generation have achieved photo-realistic results in talking-head synthesis. However, extending these models to singing-head synthesis remains challenging due to a significant Domain Gap: singing requires more exaggerated expressions, vivid jaw openings, and precise rhythmic synchronization. Current models, primarily trained on speech datasets, often struggle with "rhythmic drift" and constrained dynamics. To address this, we introduce Hi-Singers, the first large-scale, high-quality, in-the-wild video dataset specifically tailored for singing head synthesis. Hi-Singers undergoes a rigorous automated and manual filtering pipeline, ensuring strict thematic adherence, high-resolution rendering, and stable motion, resulting in 29,608 video segments totaling approximately 170 hours. We further establish a dedicated evaluation benchmark balanced across linguistic and musical styles. Extensive experiments across diverse architectures, including 3D-coefficient and diffusion-based models, demonstrate that Hi-Singers consistently and significantly improves performance across all dimensions. Specifically, it enables models to achieve superior visual realism, enhanced lip-sync consistency, and more precise rhythmic dynamics, effectively bridging the domain gap and setting a new performance standard for the singing synthesis task. The dataset is available at https://huggingface.co/datasets/CharlesZhang-USTC/Hi-Singers
Read the full abstract
Differential Geometry
A Note on Partially Anti-invariant Submanifolds of Kenmotsu Manifolds
In this paper, we introduce and investigate partially antiinvariant submanifolds of Kenmotsu manifolds. The integrability of the involved distributions is examined, and conditions ensuring that the induced foliations are totally geodesic are obtained. We further explore the geometric properties of these distributions when the associated structure tensors N and P are parallel. By making use of the curvature tensor of Kenmotsu manifolds and the properties of Kenmotsu space forms with constant psi sectional curvature, we derive various geometric characterizations of partially anti-invariant submanifolds. In addition, we characterize totally umbilical partially anti-invariant submanifolds and establish an inequality for partially anti invariant submanifolds of Kenmotsu space forms, including the characterization of the equality case.
Read the full abstract
Signal Processing
Large language models in medical time series analysis
Medical time series (MedTS), including electrocardiograms (ECG), electroencephalograms (EEG), photoplethysmography (PPG), and vital-sign recordings, are central to clinical diagnosis and health monitoring. As large language models (LLMs) have advanced, a growing body of work has examined how their reasoning, generation, and knowledge-integration capabilities can support MedTS analysis. Yet existing studies remain scattered, and the field still lacks a clear view of how these models should be designed, integrated into clinical workflows, and evaluated. This review synthesizes recent work on large language models for medical time series analysis (MedTSLLMs), covering both methodological progress and issues related to real-world deployment. We review model architectures, data resources, and processing pipelines, and prompt design strategies adapted for diverse clinical scenarios. We further organize existing MedTS applications, ranging from diagnostic interpretation and report generation to longitudinal health monitoring and physiological signal synthesis, highlighting task-specific design choices, common evaluation protocols, and empirical findings reported across studies. By bringing together current practices and open challenges, this review aims to provide a clearer foundation for developing, evaluating, and deploying MedTSLLMs responsibly in healthcare. We also maintain a regularly updated list of MedTSLLM studies and resources at: https://github.com/hy727/MedTSLLM-Review.
Read the full abstract
Plasma Physics
Scalable Production of Lead-212 and Actinium-225 Generators with Fusion Neutrons
Targeted alpha therapy will require a large expansion of ${}^{212}$Pb and ${}^{225}$Ac production. We propose neutron- and photon-driven routes that convert ${}^{230}$Th, ${}^{231}$Pa, ${}^{232}$Th, and ${}^{237}$Np into generator parents ${}^{228}$Th and ${}^{229}$Th. The most direct ${}^{225}$Ac route is ${}^{230}$Th(n,2n)${}^{229}$Th. A 10 MW deuterium-tritium (D-T) neutron source irradiating thorium with a 27% ${}^{230}$Th isotopic fraction accumulates about 250 Ci of ${}^{229}$Th in six months, initially enough for five million ${}^{225}$Ac dose-equivalents per year, and its 7916 yr half-life makes it useful for millennia. This requires only $\sim$2 gigawatt-days of D-T fusion operations. Thermal-neutron irradiation of tens of grams of ${}^{230}$Th produces gram quantities of ${}^{232}$U, a decades-long source of ${}^{228}$Th for millions of ${}^{212}$Pb dose-equivalents per year. Alternate routes proceed through ${}^{231}$Pa: fusion neutrons produce 1 to 2 tonnes of ${}^{231}$Pa per gigawatt-year from ${}^{232}$Th, and a thermal reactor converts up to 0.4 g of ${}^{232}$U per gram of ${}^{231}$Pa. Each gram of stored ${}^{232}$U produces 10 mg (8.2 Ci) of ${}^{228}$Th per year. Subsequent ${}^{228}$Th(n,$\gamma$)${}^{229}$Th converts 0.07 to 0.34 g of ${}^{229}$Th per gram of ${}^{228}$Th. We also analyze the ${}^{237}$Np(n,2n)${}^{236\mathrm{m}}$Np route to ${}^{228}$Th. These pathways scale from kilowatt to megawatt D-T sources and could secure millions of ${}^{212}$Pb and ${}^{225}$Ac dose-equivalents per year.
Read the full abstract
General Relativity and Quantum Cosmology
Noether and Mei symmetries in static spherically symmetric quadratic gravity: variational consistency, constraints, and conserved curvature flux
The symmetry content of static spherical $f(R)$ gravity is reconsidered for the pure quadratic model. We first derive the radial action without prematurely eliminating the equation carried by the radial metric variable. The Schwarzschild-type gauge may then be imposed while retaining its associated gravitational constraint. For the resulting system, radial translations and one combined scaling are exact off-shell Noether symmetries. The scaling charge reduces, on the constraint surface, to a conserved radial flux of the scalar curvature. The same flux follows independently from the trace of the four-dimensional field equations. Noether invariance, the strong point-Mei criterion, and Lie invariance of the Euler–Lagrange system are examined separately. Within the full polynomial point ansatz of total degree at most two, the Lie algebra contains precisely three independent generators. The flux also separates the solution space into a nonzero constant-curvature Einstein sector, a degenerate scalar-flat sector, and a dynamical-curvature sector. In particular, the scalar-flat sector permits a Reissner–Nordström-form metric, although its inverse-square coefficient has no electromagnetic meaning in the absence of a Maxwell field.
Read the full abstract
Neural and Evolutionary Computing
Joint UAV Activation and Placement for Post-Disaster Wireless Restoration via a Hybrid Quantum-Inspired Evolutionary Framework
In post-disaster environments, the failure of terrestrial communication infrastructure necessitates the rapid deployment of unmanned aerial vehicles (UAVs) as aerial base stations to restore wireless connectivity. This paper addresses the joint UAV activation-and-placement problem in continuous space, with the objective of minimizing the number of deployed UAVs while satisfying coverage and minimum-separation constraints. To solve this problem, we propose a Hybrid K-means Quantum-Inspired Evolutionary Algorithm (HKQEA) that combines K-means-guided initialization, a calibrated penalty-based feasibility objective, non-elitist evolutionary search, and a quantum-inspired learning update. Experimental results over 50 independent runs show that HKQEA attains a best fully feasible solution with 8 UAVs, while achieving average values of 98.94% for coverage, 99.94% for non-overlap, and 99.68% for minimum-distance satisfaction. Comparative evaluation against standard Non-dominated Sorting Genetic Algorithm II (NSGA-II), Particle Swarm Optimization algorithm (PSO) and an elitist variant of HKQEA further shows that the proposed method provides a more favorable balance among exploration, convergence behavior, and reliable feasibility preservation in constrained deployment problems. An illustrative procurement-level cost analysis also indicates that reducing the fleet from 10 UAVs to 8 can yield a 20% reduction in hardware count, corresponding to a simplified savings ratio of 25% for the studied deployment setting. These results demonstrate the potential of the proposed framework for resource-efficient post-disaster communication restoration.
Read the full abstract
Atmospheric and Oceanic Physics
A Physics–ML Multi-Fidelity Strategy for Earth System Model Parameter Optimization: A QG Proof-of-Concept
Earth System Models rely on tunable subgrid-scale parameterizations, but optimizing these parameters is computationally expensive, particularly when nonlinear interactions require many simulations. We present a hybrid Physics-ML multi-fidelity framework that combines Green's Function Optimization (GFO) with Gaussian Process or Neural Network surrogate optimization. Using a quasi-geostrophic turbulence model, GFO first ranks parameter sensitivities in normalized coordinates and selects a reduced active subset. Nonlinear surrogates then explore this subset using inexpensive 30-day simulations before refining promising candidates with 180-day simulations. Across seven strategies and a 35-member ensemble, GFO-MultiGP and GFO-MultiNN achieved mean improvements of 64.6 percent and 65.2 percent, respectively, while reaching practical saturation after 3,060 and 2,520 simulation-days. The corresponding standalone GP and NN achieved 61.1 percent and 42.0 percent improvements and required 7,740 and 6,660 simulation-days. These results demonstrate an end-to-end sample-efficiency advantage for the tested hybrid pipelines. Because screening, dimensionality reduction, initialization, and fidelity scheduling change simultaneously, their individual contributions are not isolated.
Read the full abstract
Software Engineering
Abstention Errors and Segment Support in CUAD: An Auditable Contract-Extraction Case Study
Contract-clause extraction benchmarks measure reference recovery, but interpreting a system for review assistance also requires measuring unnecessary output and the evidence available within categories. This paper presents a retrospective, reproducible evaluation of a fixed public RoBERTa checkpoint on CUAD's published 102-contract test split. Candidate generation is preserved from an earlier frozen run; matching errors are corrected and category thresholds are reselected on 62 training-split contracts using the original 90% recall target. The corrected operating point recovers 82.2% of reference spans at 8.2% CUAD precision. Of 30,464 returned candidate strings, 19,562 occur on questions with no annotated answer. The system answers 1,335 of 2,938 such questions, a 45.4% false-positive rate (95% contract-bootstrap interval: 43.3-47.9%). Matching candidates constitute 20.3% of returned strings, illustrating why reference-based precision and candidate-level review burden require different denominators. Only nine of 41 categories have at least 30 positive and 30 negative contracts; this count changes to 18 and two under support minima of 20 and 40. Two categories have no negative contracts, making their no-answer false-positive rates undefined. These findings describe one checkpoint, decoder, and threshold policy. Earlier test access and uncertain checkpoint training overlap limit confirmatory interpretation. The paper supplies the original and corrected analyses, predictions, and software checks as ancillary artifacts; it establishes neither a new release criterion nor production readiness.
Read the full abstract
Soft Condensed Matter
A thermomechanical framework for strongly nonlocal continua
The paper presents a consistent thermomechanical framework for strongly nonlocal continua, a type of generalized media in which a material's response at a point depends on deformation and temperature gradients within its neighborhood. Such dependence is responsible for localization phenomena having an intrinsic size, and nonlocal modeling also serves as a practical regularization tool to prevent mesh dependence. Nonlocality is of the integral type, using a kernel function that gives the relative influence of neighboring points on a central point. The paper develops three sources of nonlocality: (1) nonlocal momentum balance equations; (2) nonlocal conservation laws (including the first and second laws of thermodynamics); and (3) a material's nonlocal constitutive structure. Balance equations are derived using the principle of virtual power, permitting non-smooth force and displacement fields and non-smooth boundary surfaces with indistinct normal directions. The balance equations differ from those of classical local continua. Conservation laws also differ from those of local continua, with the mechanical power and heating at a point being averaged over its neighborhood. These laws are developed for processes that are sufficiently slow to minimize additional meso-scale kinetic energy due to internal turbulence. Stress, entropy, and dissipative forces are obtained as averaged derivatives of the free energy function with respect to strain, temperature, and internal variables. An example is presented of a stretched elastoplastic bar with a small defect, and nonlocality is shown to impart a characteristic size to the deformation pattern.
Read the full abstract
General Relativity and Quantum Cosmology
Survey of Novel Deep Learning Architectures for Denoising Gravitational-wave Signals
Gravitational-wave denoising must handle the full diversity of spinning, precessing binaries, since the recovered waveform underpins parameter estimation, tests of general relativity, and population studies. Matched filtering achieves this at a cost that becomes prohibitive as next-generation detectors push event rates higher; deep learning offers real-time reconstruction, but current methods are developed on narrow parameter spaces, precluding principled comparison and reliable deployment. We present the first controlled comparison of five neural-network architectures for gravitational-wave denoising, trained identically across the full astrophysically-motivated spinning binary-black-hole parameter space. A unifying principle emerges: matching network structure to the spectral anatomy of a coalescence – inspiral, merger, ringdown – outperforms brute-force model scaling. Our Multi-Scale Frequency-Aware architecture embodies this via dedicated parallel branches per frequency regime, achieving the best fidelity while using fewer parameters than larger models. It generalizes from simulated training to real LIGO-Virgo-KAGRA data without retraining, recovering merger morphology across confirmed events spanning three observing runs and both detectors, despite training on a single detector's simulated noise. We construct population-level uncertainty bands from denoising residuals, validate their calibration, and stress-test the framework on extended mass ratios and a spin population resembling hierarchical-merger remnants. Applied to real noise with no known signal, the network suppresses its output almost everywhere, with rare exceptions traced to noise transients rather than a general weakness – indicating the statistic discriminates signal from noise and motivating a future detection study. Released weights give a deployable, reproducible benchmark for future extensions.
Read the full abstract
Computer Vision
CANAL: Channel-Aware Noise Allocation for Differentially Private Feature Distillation in Medical Image Segmentation
Medical image segmentation needs diverse training data, but hospitals hold complementary scans they cannot share for privacy and regulatory reasons. Knowledge distillation can bridge this gap by exporting learned feature representations instead of images, but those representations still encode patient-specific anatomy and remain vulnerable to membership-inference and feature-inversion attacks. Adding calibrated Gaussian noise restores a differential-privacy guarantee, yet three issues have been overlooked. First, prior DP feature-distillation pipelines re-sample noise at every student iteration, so each patient image is released many times and the privacy cost composes over those releases, growing by orders of magnitude. We present a sample-once-per-image release, realized by a single precomputation pass, under which each patient contributes one release. Second, uniform noise is wasteful because channels differ in task importance. Using task-gradient energy as the importance measure, we derive CANAL, a closed-form water-filling allocation that gives important channels proportionally less noise, and prove it strictly minimizes importance-weighted distortion at a fixed budget. Third, the clipping caps and importance scores that drive the allocation are themselves data-dependent, so releasing them in the clear silently breaks the guarantee. We give a DP-honest budget split that charges each to the privacy budget, so the reported epsilon is the true epsilon. Across three medical segmentation benchmarks spanning dermoscopy, colonoscopy, and ultrasound, CANAL retains more task-relevant signal than uniform noise at the same privacy budget.
Read the full abstract
Robotics
Conflict-Predictive Variable Horizons in Multi-Drone Distributed Model Predictive Control
In distributed model predictive control for multi-drone collision avoidance, a fixed prediction horizon forces a compromise: a short horizon is inexpensive but reacts late to approaching neighbors, whereas a long one anticipates conflicts at a per-step cost that grows superlinearly with its length. We propose a conflict-predictive variable horizon that each drone sets locally, leaving the distributed model predictive control itself unchanged. From a short history of observed positions, a drone extrapolates the flight lines of its neighbors, tests each against its own using confidence funnels that narrow with prediction range, and obtains each time to conflict in closed form. The horizon is then the smallest admissible value whose planning window covers the farthest predicted conflict. It collapses to its minimum in clear airspace and grows only when a conflict lies ahead. Provided this minimum meets a single computable feasibility bound, we prove that recursive feasibility and asymptotic stability are preserved for every horizon the policy can select. These guarantees hold for a linear model, and a cascaded inner loop reduces each quadrotor's translational dynamics to a perturbed double integrator, so they carry over to the linearized quadrotor model and, as practical stability, to the full nonlinear one. In simulation on dense antipodal-swap benchmarks, the variable horizon reduces both per-step solver cost and total computation well below those of a long fixed horizon, and it maintains separation in every run, which a short fixed horizon of comparable per-step cost does not.
Read the full abstract
Computer Vision
Frame-Synchronous Hand Gesture Detection by Projected Winding Order
Gesture recognition on video is normally posed as classification: label each frame, then act on the label. That is adequate for control, where a command may be obeyed several frames late without a user noticing, and inadequate for synchronisation, where an output must be aligned to the frame on which the gesture physically occurred. We take the synchronisation problem for one common movement, the rotation of an open hand about its long axis, and show that it admits an exact solution needing no classifier, no training data and no calibration. Let s be the normalised two-dimensional cross product of the two palm edges at the wrist and the outer knuckles, under the projection the camera already performs. We prove that s factorises as k(theta)cos(theta) with |k| > 0 everywhere, so s vanishes exactly when the palm is edge-on and its sign tracks the face presented to the camera. Detection is therefore a zero crossing of one scalar, which yields an instant rather than an interval, and we prove the criterion invariant to image mirroring, hand scale and handedness, all from its algebraic form rather than from the landmark estimator. A two-handed interaction uses four fingertips as a window onto a restyled version of the same scene. We give a coverage predicate that stays correct when the hands cross, where the usual triangulation does not, and derive each free parameter of the three stylisation operators from a noise statistic rather than by inspection. Against a corpus with exact ground truth the criterion detects 95% of flips with no false positive in 240 near-miss sequences, and places each within 6.7 ms on average of the instant it occurred, a sixth of the interval at which the hand is observed, against 23.0 ms for reporting the bracketing sample. Reporting an event more finely than one samples follows from treating a gesture as the zero of a continuous quantity rather than a label.
Read the full abstract
Computer Vision
Capability-Routed Visual Retrieval and Evidence Threading for Long-Context Document Question Answering
Annual reports, diligence packs, and infographic dashboards bury numbers in page images: axes, cell grids, and footnotes that OCR pipelines flatten and that page-level visual retrievers still treat as interchangeable in-context examples. We keep a frozen Qwen2.5-VL-7B-Instruct generator and a ColPali / VisRAG-Ret page index, and insert three modules. A capability-aware visual router (CAVR) tags each retrieved page as text, table, chart, layout, or mixed and mixes specialist experts before generation. Weak-to-strong page selection (WSPS) distils a frozen 7B answerability teacher into a 3B selection head so ranking is no longer a single InfoNCE score. Visual evidence threading (VET) builds layout-anchored paths of length at most three and lets the generator read the thread rather than a flat top-$k$ list. On gold-page DocVQA / ChartQA / InfographicVQA the 7B system reaches 96.3 / 90.1 / 85.4. Under the VisRAG top-3 protocol the mean generation accuracy is 62.74 versus 59.39 for the same backbone with concatenation. On MMLongBench-Doc retrieve-then-read, F1 moves from 19.2 to 22.6 and multi-page accuracy from 16.4 to 21.2. ViDoRe nDCG@5 after WSPS reranking is 83.6, with TAT-DQA financial reports at 70.4.
Read the full abstract
Computer Vision
Structure-Token Evidence-Anchored Reasoning for Scientific Chart Understanding
Scientific charts encode quantities in axes, legends, and geometric marks, yet large vision-language models still treat them as natural photographs. Visual in-context examples do not expose the coordinate frame; unconstrained chain-of-thought can name a plausible number that was never read from a bar. We present STEER (Structure-Token Evidence-anchored Reasoning), which freezes a Llama-3.2-Vision encoder and inserts three modules: a chart structure graph encoder (CSGE) that binds ticks, legend items, and marks; evidence-anchored step reasoning (EASR) that forces every arithmetic step to cite a graph node; and weak-parser strong-reasoner alignment (WPSR) that uses a specialized table extractor only as a teacher of node attributes. On ChartQA, STEER reaches 82.70 average relaxed accuracy versus 80.16 for ChartGemma and 76.40 for a LLaVA-CoT backbone trained on the same mix. Gains widen on CharXiv reasoning (33.60 vs. 29.20 InternVL Chat V1.5) and ChartQAPro CoT (40.70 vs. 37.17 Qwen2-VL-7B), where OCR shortcuts disappear. Ablations show that dropping node serialization or numeric candidate constraints undoes most of the reasoning lift.
Read the full abstract
General Relativity and Quantum Cosmology
Black Holes, Gravitational Waves and Space-Time Singularities Lemaitre Conference 2024
This editorial introduces the topical collection arising from the Lemaitre Conference 2024 - the second in a series of meetings dedicated to the scientific legacy of Georges Lemaitre - held at the Vatican Observatory in Castel Gandolfo, 17-20 June 2024, fifty-seven years after Lemaitre's death and on the eve of the centenary of his foundational 1927 paper on the expanding universe. The nineteen contributions collected here address the most pressing open problems at the interface of cosmology, gravitation, and quantum theory: the Hubble tension and the future of ${\Lambda}$CDM in light of recent DESI data; inflationary cosmology and dark-energy model-building in string theory and supergravity, together with the Swampland constraints that bound them; the observer-dependence of the quantum-cosmological wave function; the removal, or consistent crossing, of spacetime singularities - from the canonical quantization of Lemaitre's own 1933 dust model to string-theoretic pre-Big-Bang bounces and many-body constructions of singularity-free black-hole cores; primordial black holes and their gravitational-wave signatures; the search for a background-independent formulation of quantum gravity and a first-principles account of horizon thermodynamics; and the foundational status of the quantum-to-classical transition in gravitational contexts. A proposal for lunar-based cosmological observations, and a historical reconstruction of the genesis of Lemaitre's primeval-atom hypothesis, complete the collection. A recurring unifying thread was the legacy of Lemaitre himself - who first distinguished coordinate from physical singularities and coined the term "horizon" - whose pioneering vision continues to guide research at the frontier of cosmology and fundamental physics.
Read the full abstract
Computer Vision
Forward-Facing Near-Infrared Adds Little to Colour for Farm-Machinery Traversability: A Site-Disjoint Evaluation of Sensor-Dependent Spatial Leakage
Near-infrared (NIR) imaging does not consistently outperform standard color cameras for daytime agricultural traversability once spatial data leakage is eliminated. Prior benchmarks suggesting an NIR advantage used sequence-level splits that permitted spatially autocorrelated imagery into test sets, artificially inflating NIR performance, especially on difficult paddy-boundary segmentation. When evaluated across strictly held-out recording sites using the AI Hub autonomous driving corpus, none of the four tested configurations (color, NIR, a luminance control, or their fusion) reliably surpasses standard color across any of the five traversability classes. Furthermore, apparent performance gains observed at a single held-out site completely fail to replicate when the held-out site is rotated, demonstrating that the observed benefits were site-specific artifacts rather than generalizable improvements. Correcting for spatial leakage erases the apparent lead of NIR entirely rather than uniformly degrading performance across all sensor types. Consequently, adding a forward-facing NIR camera to a daylight agricultural vehicle sensor suite remains unproven, and evaluations compromised by location leakage risk distorting sensor rankings and misleading procurement decisions.
Read the full abstract
Computer Vision
BEACON: Behavior and Appearance Control for Subject-Specific Video Generation
Generating human-centric videos that preserve both visual identity and person-specific expressive behavior remains a fundamental challenge. In addition to reproducing appearance, a model must replicate the facial behaviors that characterize how a subject expresses emotion over time. However, most state-of-the-art methods condition generation on a single reference image, which contains no information about these temporal dynamics. As a result, they tend to preserve the subject's visual identity but often produce expressions with limited variation and weak subject specificity. To mitigate this issue, we introduce BEACON, a lightweight framework for person-specific video generation that produces more expressive videos by disentangling visual identity from expressive behavior. BEACON conditions generation on two complementary signals: a reference image encoding the identity and a reference video capturing subject-specific facial dynamics. By conditioning on these complementary signals, BEACON generates videos that better preserve both the subject's appearance and characteristic facial dynamics, while also supporting identity-expression transfer. Our experiments on the MEAD and RAVDESS datasets show that by fine-tuning on approximately 2,000 pairs and updating about 1% of the pretrained Wan video diffusion model, BEACON improves facial expressivity over state-of-the-art video generation methods while maintaining competitive identity preservation.
Read the full abstract
Computer Vision
Multi-View Structure-from-Motion Enables Oriented Projective Shape Analysis in Three Dimensions
Projective shape analysis provides a geometric framework for studying landmark configurations in digital images acquired by pinhole cameras. In the classical projective shape (PS) model, three-dimensional configurations ($k$-ads) are represented as points in $(\mathrm{RP}^3)^q$, $q = k - 5$. A nonparametric test is developed in Patrangenaru et al. [12], for this framework, to determine whether an object matches a design blueprint, with each configuration reconstructed from a single uncalibrated stereo pair. Such two-view reconstructions are identified only up to a 3D projective transformation, which may reverse orientation, so that the sign-blind PS summary was the only available, before oriented projective shape (OPS) was recently considered. Multi-view Structure-from-Motion (SfM) technology removes this obstruction: its bundle adjustment is identified up to an orientation-preserving projective transformation. In this paper we revisit a well-cited three-cube object, from Patrangenaru et al. [12], with $n = 8$ SfM reconstructions built in Agisoft Metashape Professional 2.3.0, and validate our implementation by reproducing the published stereo construction from the original data. This allows us to perform what is to the best of our knowledge the first three-dimensional OPS analysis, compute its extrinsic total-variance index and perform statistical inference in this novel setting. Due to the high concentration of SfM data, the OPS index is asymptotically one-half the PS index, a structural consequence of concentration rather than a property of the object. Here our blueprint hypothesis is not rejected for any of the $q = 14$ non-frame landmarks, while the SfM reconstructions are about 26 times more concentrated than the stereo ones, a substantial gain in reconstruction precision. Sample-size, photograph-count, and frame-ordering analyses support the robustness of these conclusions.