Computer Vision
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Wednesday 2 September
Modern crop breeding demands precise organ-level analysis for trait quantification, making plant point cloud segmentation (PPCS) increasingly important. However, conventional deep learning approaches rely heavily on densely annotated datasets that are labor-intensive to acquire. Unified PPCS adaptation from distribution-shifted examples with minimal additional training remains challenging. To address this, we propose PlantC2USeg, a deep transfer learning framework featuring cross-scale consistency learning to explicitly align features across spatial scales and an information-restricted decoding strategy that prevents reconstruction shortcuts and promotes robust adaptation. The resulting pre-training enables stable few-shot generalization across species and sensing conditions, while unified fine-tuning with inherited thresholds further reduces adaptation overhead. Under full supervision on Soybean3D, PlantC2USeg achieves the highest semantic IoU and instance mWCov among compared methods, at 91.91% and 94.62%. With 20 labeled samples, it leads both metrics at 89.78% and 90.27%; with only 10 samples, it retains the highest mWCov of 83.23% while achieving 83.19% IoU. Across HR3D, 10-shot transfer to tobacco, tomato, and sorghum averages 78.41% IoU and 79.42% mWCov, while 22-shot transfer to SYAU-Maize achieves the highest IoU and mRec at 92.75% and 93.51%. Furthermore, a leading category-averaged mIoU of 85.0% on ShapeNet Part demonstrates the framework's capability to handle diverse shape variations beyond agricultural domains. These results demonstrate that PlantC2USeg reduces overall adaptation effort under distribution shifts, enabling scalable plant phenotyping and transferable 3D representation learning beyond agriculture.
Computation and Language
User Feedback Provides a Unique Signal that LLMs Can not Detect
Harnessing naturally occurring feedback from user interactions offers a promising learning signal for Large Language Models (LLMs). However, recent studies suggest this feedback is inherently noisy and difficult to leverage effectively. We challenge this conception by demonstrating that user feedback is a highly actionable signal for improvement, and that its perceived ineffectiveness stems from a systematic bias in current evaluation paradigms. To isolate the usefulness of feedback, we construct synthetic data with a definitive ground truth, alongside naturalistic data to validate that our findings hold in real-world scenarios. By comparing model revisions generated with and without access to feedback across both settings, we show that feedback-informed revisions resolve targeted issues at significantly higher rates than baseline revisions. Finally, we expose the root of the evaluation bias: when a model successfully fixes an issue exclusively due to feedback, LLM judges frequently fail to identify the genuinely corrected response, systematically preferring inferior baseline outputs instead.
Read the full abstract
Commutative Algebra
Localized Persistent Commutative Algebra
We develop a localized persistent theory of commutative algebra for Stanley-Reisner rings, based on local cohomology supported at a coordinate prime rather than at the maximal ideal. The construction is modeled on the persistent Stanley-Reisner theory of Suwayyid and Wei (arXiv:2503.23482) and its functorial development for graphs and hypergraphs (arXiv:2512.17619), in which invariants of the face ring such as graded Betti numbers and f- and h-vectors are persisted across a filtration. That framework is built from the minimal free resolution and is thus Tor-theoretic; we work instead on the injective side, and the resulting modules record information localized at a single vertex, complementing the global picture given by maximal-support local cohomology. For a vertex prime $p_i = (x_j : j \neq i)$ we prove an exact $\mathbb{Z}^n$-graded decomposition of $H^q_{p_i}(k[\Delta])$ into the maximal-support local cohomology of the deletion and of the link of the vertex $i$, the first in $x_i$-degree zero and the second repeated in every positive $x_i$-degree; at the level of graded dimensions this recovers the vertex-prime case of Rahimi's bigraded formula. With Hochster's formula this yields a closed combinatorial description of every multigraded piece. Building on this structure we introduce per-vertex persistent local cohomology numbers, prove a persistent links-Hochster formula, obtain interval decompositions of the resulting reversed-arrow persistence modules and a bottleneck stability theorem, retain multiplication by the uninverted variable as a morphism of persistence modules that the two barcodes alone do not determine, and extend the theory to an arbitrary coordinate prime, where the multiplication maps of the uninverted variables assemble into a commuting Boolean diagram of persistence modules.
Read the full abstract
Instrumentation and Methods for Astrophysics
High performance Stingray
X-ray astrophysical objects show variability on a wide range of timescales, i.e. from fractions of second to years. The open-source Python library stingray is able to perform time series analyses with a focus on high-energy astrophysics. Comprising the most commonly used Fourier analyses techniques, it also supports a range of additional extensions able to analyse pulsar data, simulate data sets and perform statistical modelling. With the latest release of the code, new Fourier methods and support for additional missions have been implemented, making Stingray more easily adaptable and extendable to other use cases. In this paper we focus on testing the performance and robustness of the latest version of stingray . We consider the possible different behaviour of the code when dealing with data sets smaller or larger than the RAM. We address the problem of dealing with large data sets and implemented parallel versions of the slowest methods in this regime. For small data sets, we investigate the possibility of a porting in GPU of key functions of the code.
Read the full abstract
Earth and Planetary Astrophysics
The Ultraviolet Spectrograph on ESA's Jupiter Icy Moons Explorer Mission (JUICE-UVS)
The Jupiter Icy Moons Explorer (JUICE) mission, led by ESA, has an Ultraviolet Spectrograph (JUICE-UVS) contributed by NASA and built at Southwest Research Institute. JUICE-UVS is designed to provide a diversity of measurements to further our understanding of the potential habitability of icy ocean worlds at Jupiter and to study Jupiter and the Jovian system as an archetype for gas giants. JUICE-UVS observes photons in the 50-204 nm wavelength range at moderate spectral and spatial resolution along a 7.5 deg slit composed of 7.3 deg x 0.1 deg and 0.2 deg x 0.2 deg contiguous sections. JUICE-UVS performs a comprehensive study of icy satellite atmospheres, plumes, surfaces, and local space environments; Jupiter's atmosphere and aurora; Io and its Io Plasma Torus; and other Jupiter system targets (rings, small moons, etc.) as available. The variety of observational techniques employed include: nadir push-broom imaging, disk scans, limb stares, stellar and solar occultations, Jupiter transit observations, and neutral cloud/plasma torus stares and scans. This paper describes the UVS investigation's science plans, instrument details, concept of operations, and data formats in the context of the JUICE mission's habitability and Jupiter system goals.
Read the full abstract
Optimization and Control
Improved Gradient Descent Lower Bounds Beyond Nesterov
We study how far gradient descent (GD) can be accelerated by predetermined stepsizes in smooth convex optimization. Going beyond the classical $\Omega(n^{-2})$ first-order oracle lower bound of Nemirovsky and Yudin (1983), we prove an $\Omega(n^{-1.6342})$ non-anytime lower bound and an $\Omega(n^{-1.2408})$ anytime lower bound. These improve the recent $\Omega(n^{-1.932})$ non-anytime lower bound of Ma and Chen (2026) and the $\Omega(n^{-4/3})$ anytime lower bound of Tsai et al. (2026), respectively. Both results continue to hold when the stepsizes may be negative. Our anytime lower bound also shows that the $O(n^{-\log_2(1+\sqrt{2})})$ rate of non-anytime silver schedules (Altschuler and Parrilo, 2025; Grimmer et al., 2025) is unattainable in the anytime setting. This establishes a strict separation between the two settings.
Read the full abstract
Computer Vision
MuyBridge: Mobile Human Center-of-Mass Estimation from Monocular Video via Sparse Fusion
The 3D center of mass (CoM) is a primary quantity in the biomechanical analysis of sport, rehabilitation, and clinical movement, yet existing 3D pose tracking, mesh recovery, and multi-view triangulation methods either optimize 3D keypoint accuracy without anatomical constraints or carry compute and capture infrastructure too heavy to deploy where CoM tracking is most useful. As a result, the metric CoM remains difficult for coaches and movement analysts to measure from a single camera where athletes train and compete. In this work, we introduce MuyBridge, an on-device system that estimates the athlete's segmental center of mass trajectory from a single phone camera video stream. MuyBridge couples a compact 2D pose network and a distilled single-step monocular depth network through an analytic metric fusion that uses anatomical and physical priors to anchor the metric CoM, requiring no 3D or task-specific supervision. Evaluated on the athletic movements of AthletePose3D (running, track and field, and figure skating), MuyBridge achieves 33-41 mm vertical CoM error and 2.3-6.6% absolute-relative range error (AbsRel) under a one-time calibration, and produces CoM estimates at the 63 FPS pose-estimation rate using asynchronous 2.86 Hz depth updates on iPhone 15. Code is available at: https://github.com/Abradshaw1/Muybridge
Read the full abstract
High Energy Astrophysical Phenomena
LHAASO-WCDA observed a $\sim$ 5 days TeV-delayed flaring event in blazar 1ES 1959+650
We report a day-scale hard lag between GeV and TeV $\gamma$-ray emission from the HBL 1ES~1959+650 in early 2024. Since the LHAASO-WCDA real-time monitoring system began operation in late 2023, multiple TeV flares from this source have been triggered, including the 1st trigger flare on 2024 February 9. A Bayesian-block analysis of the WCDA light curve identifies three TeV flares in 2024. For the second triggered flare, a discrete cross-correlation analysis reveals a $>3\,\sigma$ correlation (relative to uncorrelated red-noise simulations) at a time delay of $\Delta t = 5.0_{-2.1}^{+2.1}$ days, with the TeV emission lagging the GeV. Time-resolved spectroscopy shows that this flare has the softest TeV spectrum among these flares (intrinsic spectral index $\Gamma=3.16\pm0.18$), while the 1st trigger flare is harder ($\Gamma=2.48\pm0.21$). The observed five-day hard lag is difficult to reconcile with a purely cooling-driven temporal ordering and is consistent with scenarios in which particle energization and/or transport may contribute to the evolution. However, the current data do not uniquely identify the underlying mechanism.
Read the full abstract
Machine Learning
The Implications of Linguistic Illegibility for LLM Security
LLMs are trained to generate natural language. However, various strands of evidence indicate that an LLM's externalized linguistic outputs and mechanistically-extracted linguistic features can be an unreliable lens for understanding internal model computation. We introduce the term “linguistic illegibility” to broadly refer to scenarios in which an LLM's externalized or mechanistically-probed language artifacts fail to represent how the model actually thinks. We argue that the specter of linguistic illegibility is unavoidable for LLMs whose internal computations are not directly expressed via language, but rather math over activation spaces (with lossy translations between activation spaces and natural language happening at the bookends). If linguistic illegibility is always possible, then security mechanisms that rely on a model's linguistic self-reporting (e.g., chain-of-thought monitoring, constitutional self-critique, activation probing for linguistically-defined feature vectors) can never be completely sound; the model sandbox will always need isolation techniques whose guarantees do not depend on reading a model's linguistic state at all. We argue that observing a model's outputs using taint tracking is a promising approach for an effective sandbox: regardless of how a model linguistically self-reports, a taint tracking policy can define, a priori, various pieces of system state that should never be influenced by model-produced data. We also discuss several additional sandboxing mechanisms (e.g., robust virtualization, third-party auditing of sandboxing configurations) which collectively provide a critical floor beneath linguistic monitoring, and would have mitigated recent sandbox exploits by frontier models.
Read the full abstract
Data Structures and Algorithms
Almost Linear 3-Spanners of Temporal Cliques
Temporal graphs model dynamic networks by assigning positive integer time labels to the edges, while information propagates along temporal paths, whose edge labels are traversed in nondecreasing order. A temporal $\alpha$-spanner of a temporal graph with $n$ vertices is a temporal subgraph that approximates the minimum-hop temporal distance between every pair of vertices within a factor of $\alpha$. While general temporal graphs may not admit sparse temporal $\alpha$-spanners for any value of $\alpha$, temporal cliques are known to admit temporal $(2k-1)$-spanners of size $\widetilde{\mathcal{O}}(kn^{1+1/k})$ for every positive integer $k$. We present a simple recursive algorithm that computes, for every temporal clique on $n$ vertices, a temporal $3$-spanner of size $n^{1+2/\sqrt{\ln n}}=n^{1+o(1)}$, thereby improving the previous best upper bound of $\widetilde{\mathcal{O}}(n^{3/2})$. We also show that a modified version of our algorithm computes temporal $3$-spanners of size $\mathcal{O}(nL)$ when the lifetime is bounded by $L$, i.e., all time labels are in $\{1,\ldots,L\}$, thus improving the previous bound of $\mathcal{O}(2^Ln\log n)$. Both results are particularly striking in light of the known lower bound of $\Omega(n^2)$ on the size of temporal $2$-spanners, which already holds for temporal cliques of lifetime $L\geq 3$. Both algorithms rely on a new simple recursive decomposition that certifies temporal connectivity for a large collection of source-target pairs using only $\mathcal{O}(n)$ carefully selected edges and recursively processes only the remaining pairs. Besides yielding substantially improved upper bounds, this approach is significantly simpler than previous constructions.
Read the full abstract
Combinatorics
Chern flow and Chern moment algebras
We construct realizable-volume models over every field for the factorial normalizations of homogeneous Lascoux, Lascoux-atom, and positive Grothendieck packets, including their minimal homogenizations and layers. In particular, the construction realizes the factorial normalizations of all Schubert and key polynomials and of the minimal sign-corrected homogeneous Grothendieck polynomials. The normalized polynomials are Lorentzian, and the ordinary supports are the lattice points of integral generalized polymatroids. On a Bott–Samelson tower, row and co-row filtrations assemble the local factors into globally generated bundles; a creation-state graph absorbs the remaining kernel factors by Chern flow. We also construct intrinsic algebras of joint Chern moments. Positive inverse-Chern presentations give these algebras Hard Lefschetz and Hodge–Riemann relations, and supply source-level Hodge completions of the packets. For globally generated tropical toric bundles in the sense of Kaveh–Manon, finite generating witnesses and matroid duality provide the presentations required by Larson–Partida's theorem, without representability. These constructions yield joint Chern-number inequalities, nonvanishing polymatroids, and equality criteria.
Read the full abstract
Machine Learning
Post-Training Language Models for Gold-Medal Performance in Coding Competitions
Competitive programming has become a key test of large language model reasoning, with international competitions such as IOI and ICPC representing its most challenging settings. We present an end-to-end specialization pipeline combining large-scale problem curation, synthetic reasoning traces, supervised fine-tuning (SFT), and reinforcement learning (RL). Using 22,000 curated problems, we train Nemotron-3-Nano-CC (30B-A3B) with SFT and RL and Nemotron-3-Ultra-CC (550B-A55B) with SFT alone. We further introduce GenCorrect, a feedback-driven test-time compute strategy that iteratively generates, evaluates, and refines diverse solutions. On IOI 2025, Nano-CC improves from 130 points to 291 after post-training and to 468 with GenCorrect, exceeding the gold threshold of 438.3 while Ultra-CC reaches 502. Guided by these results, we develop a competition-specific Ultra-CC system and evaluate it prospectively during IOI 2026. Under the same time, internet-access, and submission constraints as human contestants, it scores 535.4 out of 600, exceeding both the gold threshold of 361.12 and the top human score of 498.27. To our knowledge, this is the first AI system to outscore the highest-scoring human contestant on an IOI problem set.
Read the full abstract
Differential Geometry
The universal moduli space of non-degenerate $\mathbb{Z}/2\mathbb{Z}$ harmonic spinors
This article equips the universal moduli spaces of non-degenerate $\mathbb{Z}/2\mathbb{Z}$ harmonic spinors and eigenspinors with tame Fréchet manifold structures and proves that their natural projections to the corresponding parameter spaces are uniformly Fredholm. A unified construction yields the deformation theories of harmonic spinors and eigenspinors in dimension three and of chiral harmonic spinors in dimension four, recovering earlier work of Donaldson, Parker, and Takahashi. As an application, a deformation theory for non-degenerate $\mathbb{Z}/2\mathbb{Z}$ harmonic self-dual $2$-forms in dimension four is established.
Read the full abstract
Computer Vision
RoGe: Novel View Synthesis via End-to-End Implicit Reconstruction and Generation
Novel view synthesis from sparse inputs requires both geometric grounding from the observed views and generative priors of unobserved regions, motivating recent hybrid methods that combine reconstruction and generation. However, existing methods bridge the two with rendered images or explicit 3D representations such as point maps or 3D Gaussians. Generation is thus conditioned on a lossy and imperfect projection of the scene, inheriting its errors, and reconstruction receives no signal from generation to correct them. We present RoGe, an end-to-end unified reconstruction and generation framework that removes this explicit bridge. It targets roaming within a scene anchored by sparse views: given a few posed images and a camera trajectory, it synthesizes a temporally coherent video along that trajectory. From the sparse input views, RoGe builds an implicit scene representation with a feed-forward reconstruction model, and queries it with camera rays to obtain per-view geometric features. These features are injected into a video diffusion model as conditioning, without any explicit 3D intermediate. Both modules are trained jointly, so the generation objective directly shapes its own geometric conditioning. We conduct extensive experiments, where RoGe outperforms reconstruction-based, generation-based, and hybrid baselines in terms of image-level quality and video-level temporal and geometric consistency. Ablations confirm that ray-queried implicit features outperform both raw reconstruction tokens and rendered RGB as conditioning, and that joint training brings further gains. Our code will be released on https://jerry-locker.github.io/roge/.
Read the full abstract
Machine Learning
UE5M3 FP4 Block Scaling for Stable Language Model Pretraining
Stable 4-bit floating-point (FP4) pretraining is difficult because the E2M1 payload represents only a narrow range of magnitudes. NVIDIA's Transformer Engine \nv{} recipe addresses this with current-tensor scaling, a randomized Hadamard transform (RHT), and bfloat16 (BF16) final layers, adding work outside the FP4 matrix multiplications. We instead pair E2M1 payloads with unsigned E5M3 (\ue{}) block scales. Their wider range permits periodic tensor scaling, while our recipe applies selective stochastic rounding to backward gradients, omits RHT, and uses FP4 in all eligible internal linears. We pretrain a Nemotron-H 8B model for nearly 190 billion tokens. Compared with Transformer Engine \nv{}, the proposed block-16 recipe finishes with lower final-window training loss and, under their respective quantized-inference policies, lower validation loss measured as held-out negative log-likelihood. Its quantized-inference downstream point estimates are also higher on all three reported aggregates. A native \nv{} execution ablation that jointly removes RHT and the BF16 final-block exemption increases measured model-body token throughput by 21.2%. These results demonstrate end-to-end software-emulated \uefp{} pretraining with a simpler recipe and motivate native support for \ue{} block scaling.
Read the full abstract
Combinatorics
Degeneracy bounds, stability, and a sharp gap for $B$-colorings
A $B$-coloring of a graph is a proper edge-coloring in which every $4$-cycle is rainbow, and $q_B(G)$ denotes the minimum number of colors in such a coloring. Let $\Delta_2(G)$ denote the maximum number of common neighbors of two distinct vertices of $G$. We prove that, for integers $1\le d\le\Delta$, every finite simple $d$-degenerate graph $G$ with $\Delta(G)\le\Delta$ satisfies $$q_B(G)\le \Delta+(d-1)\Delta_2(G)\le d\Delta.$$ Consequently, $d\Delta$ is the exact maximum, with equality precisely for graphs containing $K_{d,\Delta}$. More generally, if $q_B(G)\ge d\Delta-s$, where $0\le s<\Delta$, then $G$ contains $K_{d,\Delta-s}$; if also $s<d$, then $G$ has at least $d-s$ vertices of degree $\Delta$ with the same open neighborhood. For $\Delta\ge3$, we further show that every $K_{3,\Delta}$-free 3-degenerate graph satisfies $q_B(G)\le3\Delta-2$; the example $K_{3,\Delta-1}$ shows that this bound is best possible up to one. For loopless multigraphs, we establish a sharp gap in the possible values of $q_B(G)$. For every integer $\Delta\ge3$, every finite loopless multigraph $G$ with $\Delta(G)\le\Delta$ satisfies $$q_B(G)\le\Delta(\Delta-1)$$ unless $G$ has a component isomorphic to $K_{\Delta,\Delta}$, in which case $q_B(G)=\Delta^2$. The bound $\Delta(\Delta-1)$ is attained by both $K_{\Delta,\Delta-1}$ and $K_{\Delta,\Delta}-e$. Consequently, among finite loopless multigraphs with maximum degree at most $\Delta$, no value of $q_B(G)$ lies strictly between $\Delta^2-\Delta$ and $\Delta^2$.
Read the full abstract
Exactly Solvable and Integrable Systems
Inverse scattering method for nonlinear negative first order coupled Klein-Gordon equation
The exact N-soliton solutions are derived for the coupled negative first order Klein-Gordon (CNKG) equation subject to vanishing boundary conditions by using the inverse scattering method via Gelfand-Levitan-Marchenko equation. Based on the zero-curvature representation, the conservation laws, integrals of motion, and Hamiltonian structure of the aforementioned coupled nonlinear equations are constructed. The Jost functions and their analyticity properties for the Manakov spectral problem are recalled. The integral equations for the eigenfunctions are then used to formulate the Gel'fand-Levitan-Marchenko equations. Solving these equations establishes a direct correspondence between the kernel functions and the potential, yielding the general N-soliton expressions.
Read the full abstract
Number Theory
Petersson-Rigid Lattices in a Census of 100 Rank-Three Root Bases
The first paper worked out one family of four lattices in full – $|\det| = 12, 24, 36$ and $72$. This paper generalizes it. We enumerate the 44 symmetrizable rank-three hyperbolic Cartan matrices and their 56 depth-one edits, 100 root bases in all, and compute the obstruction space $S_{5/2}(\rho_L)$ of the 98 within our weight-$5/2$ budget: 10 vacuous, 34 unobstructed, 54 obstructed. The vacuous ten are the lattices Bruinier, Ehlen and Freitag call simple, read in signature $(2,3)$. Of their fifteen, our ten realize five; of the rest, five need more than three generators and cannot be the discriminant form of a rank-three lattice at all, three fail $|\det L| = 2k^2$, and two are simply not root bases. Within the rank-three hyperbolic world the simple lattices are exactly the Feingold-Frenkel neighbours of index $k \le 4$. The discriminant group $L'/L$ carries a finite quadratic form, and the finite group of its isometries acts on the weight-$3/2$ cusp forms for $\rho_L$, the bottom antisymmetric rung. We survey the lattices on which that action is absolutely irreducible of dimension at least two, so that the Petersson pairing there is pinned down up to a single scalar. The condition alone cuts the census to three: $L_4$ of the first paper, and two new ones at $|\det| = 40$ and $88$. The quaternion discriminants that occur are 6, 10 and 22, the three for which the Shimura curve $X^D$ has genus zero. Rigidity uses no quaternion input, so we record it as an observation, not a characterisation. Both invariants of the first paper – the shadow norm $\|\Xi\|^2$ and the Petersson scalar $t$ – were single points there. Three rigid lattices instead of one are where their special faces come off: $\|\Xi\|^2$ generalizes against $L(f,1)$ where the first paper read $L(f,2)$, and $t$ against an elliptic curve's imaginary period where it read $\Gamma(1/3)$. Both numerically, to 26-58 digits.
Read the full abstract
Analysis of PDEs
Global Well-posedness and Asymptotic Analysis of a Damped Nonlinear Wave Equation with a Codimension-One Constraint
We prove the global existence and uniqueness of strong solutions to a constrained version of the damped nonlinear wave equation $$ \vartheta_{tt}+\gamma \vartheta_t-\Delta\vartheta+|\vartheta|^{p-2}\vartheta=0 $$ on a smooth bounded domain $\varOmega\subset\mathbb{R}^d$, where the evolution is projected onto the tangent space of the Hilbert manifold $$\mathcal{M} = \left\{ \vartheta\in L^2(\varOmega):\|\vartheta\|_{L^2(\varOmega)}=1 \right\}, $$ which is the unit sphere in $L^2(\varOmega)$. We assume that $$p\in[2,\infty)\ \text{ for }\ d=1,2, \ \text{ while }\ 2\leq p\leq \frac{2(d-1)}{d-2} \ \text{ for }\ d\geq3.$$ By employing the Faedo–Galerkin approximation method, together with suitable a priori estimates and compactness arguments, we establish the global well-posedness of the problem. In particular, we show that the Hilbert manifold $\mathcal{M}$ is invariant under the flow, and hence the $L^2$-constraint is preserved throughout the evolution. Using the Lusternik–Schnirelmann theory, we show that the corresponding stationary problem possesses at least countably many stationary solutions. We further investigate the long-time behaviour of solutions and prove that, along a subsequence, every solution of the constrained problem converges to a stationary solution by invoking Webb's theorem and Barbalat's lemma. When the initial data are sufficiently close to the first eigenfunction of the associated stationary problem, we show that the unique strong solution converges in $H_0^1(\varOmega)$ to the unique positive ground-state solution.
Read the full abstract
Quantum Physics
Exponential speedup of polarization stabilization for long distance DWDM quantum networks
Fibre-based quantum networks distributing polarization entanglement require a stable and uninterrupted transmission basis for reliable operation. Bright classical reference light enables rapid polarization feedback but can introduce noise into quantum channels. Entangled-photon-based feedback avoids this noise, but typically interrupts the target entanglement channel during calibration and becomes prohibitively slow over long distances due to the product loss of fibre links. Here, we overcome both limitations by combining wavelength-bracketed probing with switch-enabled path decomposition. Spectrally adjacent entangled-photon sidebands track the polarization response of the central distribution channel without interrupting its transmission, while optical switches and local reference fibres independently determine the signal and idler network transformations. Transferring the resulting compensation settings to the central channel eliminates calibration-induced downtime and changes the acquisition-time scaling from the product of the link losses to the sum of losses. We demonstrate the method on a 133 km fibre testbed and achieve continuous closed-loop stabilization for more than 24 hours without classical reference light. The decomposition of multi-link quantum feedback into single-link measurements provides a scalable stabilization strategy for wavelength-multiplexed quantum networks.
Read the full abstract
Lattice Field Theory
On the equivalence of left-hand-cut and three-body formalisms
The "left-hand cut problem" in lattice QCD arises when exchange singularities invalidate standard relations used to extract scattering amplitudes from the finite-volume spectrum. In recent years, several approaches were proposed to resolve this problem. We establish analytically and verify numerically the equivalence between the three-body approach proposed in [JHEP 06 (2024) 051] and the two-body left-hand cut formalism developed in [JHEP 08 (2024) 075]. We test the equivalence in a scalar model describing a particle scattering from an $S$-wave two-particle bound state through one-particle exchange. Both approaches reproduce the same infinite-volume particle–bound-state amplitude and correctly encode the associated left-hand cut in the finite-volume spectrum. Crucially, we identify that the exponentially suppressed finite-volume effects neglected in both formalisms can be numerically large, compromising amplitude extraction at small lattice volumes.
Read the full abstract
Computer Vision
Efficient All-in-One Weather Restoration using Spectral Harmonization
Adverse weather conditions such as rain, haze, and snow significantly degrade image quality, posing challenges for both human perception and physical AI. Existing restoration methods require large computational budgets, struggling to process high-resolution images and handle different degradations. In this paper, we present Frequency Reconstruction via Spectral Harmonization, a novel lightweight all-in-one restoration method that explicitly decomposes feature representations into high- and low-frequency components at each scale of a hierarchical encoder-decoder architecture. By combining spectral decomposition with spatial processing through Fourier-based skip connections, FReSH-IR captures complementary frequency information without sacrificing spatial detail. Our approach achieves similar restoration quality with 80% fewer parameters and operations than transformer-based models. Extensive experiments demonstrate that our method offers a great efficiency-performance trade-off, highlighting its practical applications in constrained-resource systems.
Read the full abstract
Systems and Control
Tests on the POD-P controller of INELFE Spain-France VSC-HVDC interconnector
INELFE interconnector consists of a 2x1000 MW high voltage direct current system based on voltage source converters (VSC-HVDC) interconnecting France and Spain. INELFE VSC-HVDC link is embedded into the high voltage alternating current (HVAC) system. Electromechanical oscillations, also known as power oscillations, are a major concern worldwide. INELFE VSC-HVDC link has specific controllers to damp power oscillations by modulating active (P)- and reactive (Q)-power injections of the VSC converters (POD-P and POD-Q controllers, respectively). The Spanish and French Transmission System Operators (TSOs) carried out a join task force to - Increase the gain of POD-P controller of INELFE VSC-HVDC link. - Make it possible to use the POD-P controller together with angle difference control (ADC) of INELFE VSC-HVDC interconnector. The objective of these modifications is to improve the effectiveness of POD-P controller and, therefore, to increase its contribution to the damping of inter-area oscillations in the Continental Europe (CE) power system. Such changes require (a) extensive simulation studies and (b) extensive tests in different operation modes, in order to ensure the correct behavior of the system. This paper presents simulation studies and field tests on the POD-P controller of INELFE VSC-HVDC interconnector in different modes of operation.
Read the full abstract
Functional Analysis
Unitarily invariant norms
This survey paper provides a comprehensive study of unitarily invariant norms on the algebra of $n \times n$ matrices. This investigation leads naturally to the theory of symmetric gauge functions, a class of norms on $\mathbb{R}^n$ characterized by invariance and monotonicity properties. We develop the necessary framework by examining absolute and monotone norms and establishing their equivalence, thereby offering additional insight into the classical Hardy-Littlewood-Pólya theorem on majorization. The theory of majorization is further explored through its connections with doubly stochastic matrices, convexity, and fundamental results such as the Birkhoff and Radó theorems, as well as König's theorem on term rank and line rank. We also study weak majorization and derive a characterization that plays a crucial role in proving the monotonicity of symmetric gauge functions. On the spectral side, we review key results in matrix analysis, including the Courant-Fischer min-max theorem, the Cauchy interlacing theorem, and Ky Fan's majorization theorem, along with a weak subadditivity result for singular values of arbitrary matrices. These ingredients culminate in a detailed proof of von Neumann's characterization of unitarily invariant norms, which provides a complete and elegant description of this class of norms. Some illustrative examples, as well as the Ky Fan domination principle as an application, are also presented.
Read the full abstract
Statistical Methodology
Non-Invariance in Nested Prediction Models under Selective Predictor Availability
We used nested models as a framework for characterising the consequences of a selectively measured predictor in clinical prediction models. In this framework, a model containing predictors available in the target population is referred to as the restricted model, while the extended model additionally includes a selectively measured predictor. We show that non-invariance in the restricted model between selected patients and the target population decomposes into components due to omission of the additional predictor, potential residual non-invariance, and their interaction. This framework is extended to a predictor measured via multiple routes of selection resulting in collider structures. We further show that imputation based on the conditional distribution of the additional predictor in the selected population transfers the restricted-model non-invariance to the imputed extended model as imputation bias. We propose diagnostic tests for selective measurement and restricted-model non-invariance that can be applied to routinely collected data, and illustrate the framework using publicly available intensive care data, in which blood lactate is selectively measured. The framework provides a formal basis for understanding how selective predictor measurement affects generalisability of clinical prediction models using routinely collected health data.
Read the full abstract
Sound
Understanding Automatic Mixing: A Subtask-Oriented Analysis of Two-Stage Mixing System
Automatic mixing transforms multitrack recordings into perceptually coherent, balanced, and aesthetically consistent mixes. In real-world production, this task is challenging due to large track counts, diverse instrumentation, and strong inter-track dependencies. Two-stage systems address this complexity by separating intra-group processing from inter-group mixing, yet it remains unclear whether their gains arise from stronger component models or from explicit task decomposition. We present a subtask-oriented analysis of automatic mixing through three controlled listening experiments. We investigate whether full-mix models transfer to intra-group mixing, whether downstream models compensate for grouping and loudness errors, and whether two-stage decomposition improves full-mix quality. Across three dense pop and rock excerpts, transfer differs between the evaluated models; inappropriate grouping causes clear downstream degradation, while altered loudness relationships have weaker and model-dependent effects. Both two-stage variants significantly outperform their corresponding single-stage baselines. These findings support explicit separation of local balance and global mix coordination as a useful design principle for automatic mixing. Code and audio examples are available online.
Read the full abstract
Combinatorics
Duals of algebraic matroids need not be algebraic
We give an example showing that the dual of an algebraic matroid need not be algebraic. Our example is a matroid of rank 6 on 10 elements which is algebraic in characteristic 2. It is obtained by gluing an algebraic realization of the non-Fano matroid to a realization of the Fano matroid. Its dual is shown not to be algebraic using the Ingleton-Main lemma.
Read the full abstract
Computational Physics
Learning Spectrally Optimised Mesh-Free Discretisations
Numerical methods for partial differential equations (PDEs) discretise differential operators, ideally reproducing the action of the continuous operators across all wavenumbers permitted by a given discretisation. Spectral-type methods approach this ideal, but rely on structured grids or high-order meshes that are hard to generate for complex geometries. In contrast, mesh-free methods are geometrically flexible, yet how faithfully they reproduce the operator across resolved scales is strongly influenced by a heuristically chosen kernel, which selects one of many weight sets satisfying the same consistency conditions without regard to the resulting spectral response. To address this, we introduce Spectrally optimised Neural Discretisations (SpeND), a framework that learns the map from local stencil geometry to discretisation weights on unstructured point clouds. A projection layer constrains every predicted stencil to the affine set defined by the discrete moment conditions, so polynomial consistency, and hence formal order of accuracy, holds exactly. Since consistency is enforced by the architecture, the weights can additionally be optimised for accuracy on a prescribed function space. Herein, we use Fourier modes and target the exact differentiation response over a chosen wavenumber band, so training is unsupervised. The loss function is a design interface: changing how it weights wavenumbers yields operators with distinct accuracy profiles. The trained discrete differential operators are PDE-agnostic and are applied without retraining to the Poisson, Burgers and Navier–Stokes equations, to near-boundary stencils, and across resolutions. At the same order and resolution, SpeND matches or improves on established mesh-free discretisations, and has been shown to reduce the wall-clock time between $3-20\times$ to achieve equivalent error as the baselines.
Read the full abstract
Machine Learning (Statistics)
Copula Transformations for Data-Consistent Inversion
Data-consistent inversion (DCI) constructs probability measures whose push-forward distributions agree with observed data, while iterative data-consistent inversion (iDCI) extends this framework to generalized stochastic inverse problems by enforcing multiple push-forward constraints sequentially. Although iDCI avoids the direct approximation of high-dimensional joint densities, its relationship to the original joint DCI solution has remained unclear. In this work, we establish this relationship through copula theory. Using Sklar's theorem, we derive a factorization of the DCI update into separate marginal and dependence transformations and show that the discrepancy remaining after convergence of the iDCI algorithm is entirely characterized by the copulas associated with the observed and predicted joint distributions. This characterization motivates a copula-transformed iDCI solution, and we prove that an exact copula transformation recovers the original DCI solution. We further establish convergence results for approximate copula transformations under converging sequences of reference measures and progressively enriched feasible sets. Numerical examples demonstrate how the geometry induced by the quantity-of-interest map governs the importance of the copula transformation, illustrate an adaptive reference-measure refinement strategy for improving computational accuracy under a fixed sampling budget, and demonstrate the progressive refinement of generalized stochastic inverse problems through heterogeneous, asynchronously acquired experiments.
Read the full abstract
Computer Vision
Benchmarking RAW and RGB Restoration in Image Signal Processors
Modern cameras transform RAW sensor measurements into sRGB images through an image signal processor (ISP). We benchmark two placements for blind restoration around a fixed ISP: (A) pre-ISP restoration in the RAW domain and (B) post-ISP restoration in the sRGB domain. The benchmark covers four smartphone device groups, two learned ISPs, three degradation regimes–noise, blur, and joint noise and blur–, and several representative RAW and RGB restoration models. Our results show that placement alone does not determine performance. The RAW restoration strategy outperforms the best generic RGB restoration models. However, RGB restoration models trained considering the ISP transformations, achieve the best overall performance. Our novel benchmark demonstrates that the image reconstruction performance strongly depends on the alignment between the restoration model and the target imaging pipeline. We consequently recommend reporting restoration placement and ISP-aware supervision as key experimental factors. Our code is available at https://github.com/mv-lab/AISP