Peer Revenue Publish or profit

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

All of arXiv

Saturday 5 September

Optimization and Control

Isolated calmness of regularized linear inverse problems

This paper studies the isolated calmness property of solution mappings arising from convex regularized linear inverse problems. We mainly establish a tangent-cone characterization of this property. For convex piecewise linear-quadratic regularizers, the isolated calmness coincides with the solution uniqueness, leading to simple verification procedures. In particular, for analysis sparsity regularization, our condition can be verified by linear programming.
Read the full abstract

All authors Tran T. A. Nghia, Huy N. Pham, Nghia V. Vo, Khoa V. H. Vu

Algebraic Geometry

Probability-theoretic interpretation of degeneracy locus formulas

Schwartz-MacPherson classes of degeneracy loci of symmetric and skew-symmetric maps. Together with the known formulas for ordinary linear maps, this completes the stable SSM theory for the three classical types. We interpret these formulas probabilistically, in terms of the endpoint random partition of a stochastic rational six-vertex model: the skew classes are probabilities of Maya-dimer events, and the symmetric classes are Chebyshev moments of the random 2-core.
Read the full abstract

Superconductivity

Effect of non-homogeneous non-magnetic Impurities in Superconductors: A comparison between Microscopic and Macroscopic theories

Non-homogeneous, non-magnetic impurities in superconductivity have recently gained traction as a promising approach to reduce surface resistance and increase the superheating field. We aim to deepen the understanding of impurities' role from a theoretical perspective by utilizing Eilenberger's equation, and to investigate the validity of a more phenomenological approach based on London's equation for determining the magnetic field profile within the superconductor. We show that the microscopic and macroscopic theories produce identical electromagnetic field distributions for any spatial distributions of impurities inside the superconductors. However, the two approaches provide different magnetic field-related quantities. The macroscopic model only gives the Bean-Livingston barrier, while the microscopic theory provides the experimentally-relevant superheating field. Based on the microscopic formalism, we determine a family of impurity profiles that make the superheating field $H_{\rm{sh}}$ equal to the critical field $H_c$, which is the maximum achievable.
Read the full abstract

All authors Carlos Redondo Herrero, Tejas Guruswamy, Orlando Quaranta, Akira Miyazaki

Logic in Computer Science

Differentiable Horn Programs: A Constructive Expressivity Theorem for Latent Rule Operators

We introduce LatentGamma, a differentiable operator on the unit cube $[0,1]^M$ designed as a smooth surrogate of Tarski's immediate consequence operator $\TP$ associated with a definite Horn program $P$ on $M$ atoms. The operator is built as a five-stage composition combining sigmoidal gating, softmax routing, and a residual update that enforces monotonicity by construction. We establish four theoretical results. First, the iterated sequence is coordinate-wise non-decreasing and bounded by $\ind$, hence converges to a fixed point of LatentGamma; the operator itself is lattice-monotone on $[0,1]^M$. Second, our constructive expressivity theorem shows that for every definite Horn program $P$ there exists a closed-form parameter assignment $\theta^*(P)$ and an explicit time bound $T_{\max}(P)$ such that the iterated sequence exactly reproduces $\TPinf(F_0)$ for every initial fact set $F_0$ throughout the window $[D(P, F_0), T_{\max}(P)]$, where $D(P, F_0) \leq M$ is the derivation depth. The window $T_{\max}$ is large in practice ($\geq 10^5$ for sparse programs) and reflects the finite-time nature of computation by smooth sigmoidal gates. Third, the oracle is robust to Gaussian noise on its body and head logits, with explicit non-asymptotic bounds. Fourth, we prove a matching information-theoretic lower bound on the parameter count. We provide a complete numerical validation on programs ranging from $M = 33$ to $M = 504$ atoms: oracle accuracy reaches $1.0000$ on $2000$ test cases with zero false-positive and false-negative rates, and the empirical noise tolerance $\sigma_{\max}$ scales precisely as the union bound $RM \cdot \Phi(-10/\sigma_\eta)$ predicts.
Read the full abstract

Combinatorics

Distance statistics of block-weighted planar quadrangulations

We study random planar quadrangulations in which each block – i.e., each simple (without multiple edges) component – is assigned a weight $u$. We derive an explicit expression for the distance-dependent two-point function, defined as the generating function of such block-weighted maps with two marked edges at a fixed graph distance. Using contour integral representations in two variables combined with a delicate saddle-point analysis, we compute the associated distance profile in the scaling limit of large quadrangulations. We recover the known phase transition at $u = 9/5$, characterized by distinct scaling exponents and different scaling functions, below, at, and above criticality. We also discuss the block distance profile, where the two marked edges are conditioned to lie within the same block.
Read the full abstract

Optimization and Control

Quantitative comparison of closed- and open-loop linear-quadratic $N$-player differential games

We compare closed-loop and open-loop Nash equilibria in a finite-horizon stochastic linear-quadratic $N$-player game with decoupled state dynamics and interaction through the state costs. We introduce a block-diagonal reference game and a nearby perturbed game, and study each under both information structures. The closed-loop and open-loop problems lead to different Riccati systems, but for the reference game their induced equilibrium state and control processes coincide exactly. We then prove solvability and stability under weak perturbations of the state costs, yielding quantitative bounds between the corresponding equilibria. In particular, when the perturbation decreases sufficiently fast with the population size, the closed-loop and open-loop equilibria of the perturbed game become asymptotically equivalent. The comparison is carried out directly at the finite-player level, without requiring exchangeability or a mean-field limit.
Read the full abstract

Computer Vision

Spatial Attention Supervision for Defect Localization: Exploiting Ground-Truth Masks as Training Signal in Diffusion-Augmented Defect Detection

Ground-truth defect masks in industrial inspection datasets are typically reserved for evaluation. This paper repurposes them as spatial supervision signals during training of classification networks, teaching a model not just what to predict but where to look. The method adds an activation-based attention alignment loss that steers convolutional feature maps toward defect regions, in a mixed-supervision formulation that also accommodates samples without masks, such as diffusion-generated images. Combined with DDPM augmentation, synthetic images contribute quantity while masks contribute spatial precision. We evaluate 85 models (four CNN backbones under a 2x2 data/training factorial over five seeds, plus a Swin-V2-T transformer baseline) on the MVTec-AD bottle benchmark, with localization measured on held-out defect images excluded from classifier gradient updates. Main findings: (1) attention-guided training improves activation-based localization (Pixel-AUROC) by +18.0% for EfficientNetB0 with augmentation (p=0.005, Cohen's d=2.6) and +18.7% for ResNet50 (p=0.008), significant in four of eight CNN settings (uncorrected for multiple comparisons) with no significant change in classification; (2) for EfficientNetB0 a data x training-mode interaction is significant (p=0.002), consistent with a super-additive effect (+13.6% combined vs +1.6% summed individual effects); (3) architectures with weaker spatial representations benefit most, whereas ConvNeXt-T shows no effect, apparently because its depthwise-convolution activations yield spatially uninformative channel-mean maps; (4) unsupervised PatchCore remains the strongest localizer (Pixel-AUROC=0.983), contextualizing the supervised gains. These results show that existing evaluation masks can act as practical training signals that measurably and reproducibly improve where defect classifiers attend.
Read the full abstract

All authors Sajjad Rezvani Boroujeni, Muskan Saraf, Gnana Tulasi Makineni, Tom Bush, Hossein Abedi

Computers and Society

Formalising and Implementing Grassroots Social Contracts: From Legal Text to Grassroots Platforms (Full Version)

Two centuries ago Pierre-Joseph Proudhon envisioned a social contract that is (1) an agreement of man with man; (2) reciprocal; (3) imposing no obligation upon the parties except that which results from their personal promise; (4) subject to no external authority; (5) freely accepted and signed by all the participants; (6) of the nature of a contract of exchange. He also envisioned a property of social contracts, analogous to a property of digital platforms we term grassroots: that one could "make a contract with all, as ... with some"; digitally, that a grassroots platform can have multiple instances, which may coalesce by mutual consent. We define grassroots social contracts as social contracts meeting these conditions and that (7) people are free to deal with each other; and (8) there is no external register of people. We show that a grassroots social contract can be transformed into a working grassroots platform through an abstraction cascade, from the contract text to formal act schemas, verified syntactically to be grassroots, and then to volition-guarded multiagent atomic transactions. Act schemas are a formal language for the acts the contract describes, each naming the parties' roles, which of them must will the act, and its precondition and effect at each role. Any contract written in this language meets the eight conditions, provided it is syntactically grassroots, satisfying three decidable conditions, Introduction, Provenance and Volition. We prove that the protocol realising a syntactically grassroots contract is grassroots, and volitionally so. Clauses of a grassroots social contract are of three legal types: a breach of an enforced clause is impossible with a correct implementation; a breach of an attested or undertaken clause can be taken to court, with signed evidence for an attested clause. We illustrate them with a grassroots social graph and grassroots currencies.
Read the full abstract

Quantum Physics

Wigner Time: from experimental intent to real-time laboratory control

Precisely timed, multi-device control in atomic, molecular and optical physics is provided by specialized real-time systems, which impose their own terms on the description of the experiment: a program over a global clock, in which every stage boundary is an absolute time computed from everything preceding it. Such descriptions are tightly coupled – changing one stage affects all later ones – and are correspondingly hard to reuse, inspect, or move elsewhere. We introduce Wigner Time, a Python package in which the experimental procedure is instead represented as data: a table of timed updates, assembled by composing functions and referred to named points in the experiment rather than to absolute instants. Hardware enters only at a final conversion step, leaving the description readable and back-end agnostic. Wigner Time has run two cold-atom setups for more than two years, where taking the computation out of the real-time program improved the achievable temporal resolution five-fold; the design applies to any domain requiring precise multi-device timing.
Read the full abstract

All authors T. W. Clark, B. Sárközi, A. Dombi, Á. Kurkó, D. Nagy, A. Simon, D. Varga, P. Domokos, A. Vukics

Software Engineering

SWE-Test: Benchmarking LLM Vulnerability Discovery via Input Prediction

Vulnerability discovery is becoming an important ability of large language model (LLM) agents: agents that silently miss real defects leave critical software exposed. Rigorously measuring this ability is therefore urgent, but existing benchmarks are gameable through data contamination, score recall against an unknowable vulnerability set, often rely on synthetic bugs, and report a single end-to-end verdict that cannot localize where an agent fails. Vulnerability discovery is a composite ability: an agent must comprehend source code, infer input constraints, construct inputs, execute them, and iteratively correct from feedback. We recast its measurement as an input-prediction task with a closed, deterministic ground truth: using coverage-guided fuzzing, we mine deep target branches in real-world C/C++ programs and ask an agent to predict an input that drives execution to a given branch. This decomposes discovery into three task modes over 22 real-world C/C++ programs spanning 15 domains. Open-loop and Feedback-enabled share 60 fixed-target task instances across 16 of these codebases (13 domains), testing input construction without and with a distance oracle to isolate code comprehension from feedback-driven correction. Online Arena instead removes the predefined target and scores path exploration by coverage gain on a separate, partially overlapping pool of 11 programs; agents collectively confirmed 13 distinct bugs across six programs. Evaluating 15 default-effort model-scaffold configurations, the best reaches only 55.0% pass rate in the Feedback-enabled mode, and the mean across seven paired Claude Code configurations is 36.4% with feedback versus 19.3% without. Decomposing failures, we find constraint inference, not navigation, is the dominant bottleneck. We release SWE-Test with a turnkey evaluation environment.
Read the full abstract

All authors Yuanxiang Shi, Jiayi Lin, Xuanyong Lin, Liangcai Su, Yeheng Duan, Wei Wang, Qi Han, Bing Zhao, Wei Hu, Xander Xu, Chenxiong Qian

Functional Analysis

Functional representation and functional calculus for controlled paths

We study the relation between non-anticipative functional calculus and compatible families of controlled paths. For a non-anticipative functional satisfying horizontal Lipschitz regularity, we show that the iterated vertical derivatives generate compatible higher-order controlled Taylor expansions along Hölder controls, with level-dependent remainder exponents. We derive a rough-integration criterion from these estimates and show that, for $\gamma-$Hölder controls with $\tfrac{1}{2}\geq \gamma>\sqrt{2}-1$, the first-order remainder estimate is recovered. Our main result is a converse representation theorem. We consider non-anticipative functionals $G_0,\ldots,G_p$ which satisfy compatible higher-order controlled Taylor estimates along $\gamma$-Hölder paths. Under natural continuity and compatibility assumptions, we prove that they can be represented as the iterated vertical derivatives of the base functional: $ G_j=\nabla_\omega^jG_0,\qquad j=1,\ldots,p.$ Thus the Gubinelli coefficients of a compatible controlled family are symmetric and uniquely determined by its base functional; in particular, the Gubinelli derivative is identified with the vertical derivative introduced in functional Itô calculus, giving the coefficient hierarchy an intrinsic path-space differential structure. We show that this class of compatible coefficient families is stable under admissible non-anticipative functional composition and derive the corresponding functional chain rule. As an application, we obtain well-posedness for a class of path-dependent rough differential equations with Volterra memory, and identify the Gubinelli derivative of the resulting path-dependent rough coefficient.
Read the full abstract

Analysis of PDEs

Notes on linear elliptic equations with $L^2$-gradient perturbations and singular zero-order coefficients

In this paper, we study the existence, uniqueness, and quantitative estimates for weak solutions to linear elliptic Dirichlet problems of the form \[ -\operatorname{div}(\gamma \nabla u)+\langle \nabla\phi+\mathbf{H},\nabla u\rangle+(c+\alpha)u=f \quad\text{ in }U, \quad\; u=0 \quad\text{on }\partial U, \] where $U\subset \mathbb{R}^d$ is bounded, $\gamma\in[1,\infty)$ is a constant, $\phi\in H^{1,2}(U)\cap L^\infty(U)$, $\mathbf{H}\in L^p(U,\mathbb{R}^d)$ for some $p \in (d, \infty)$, and $c\in L^1(U)$ with $c\ge0$. A key feature of this setting is that the drift contains the low-regularity term $\nabla\phi$, which is only assumed to belong to $L^2(U,\mathbb{R}^d)$, while the zero-order coefficient is merely integrable. We prove that, even under these rough assumptions, well-posedness and quantitative energy and $L^2$ estimates remain valid. In addition, by using a previously established interpolation result, we characterize a trade-off between the integrability of the source term $f$ and that of the zero-order coefficient $c$, and show that well-posedness together with the corresponding quantitative estimates hold under these interpolated assumptions.
Read the full abstract

Cryptography and Security

Scratchy: Visual-Scratchpad Multimodal Reasoning for Cryptographic Proof Generation in EasyCrypt

Large language models (LLMs) have recently made substantial progress in formal proof generation, yet presenting distinctive challenges in cryptographic area. Computational security arguments posit that a valid proof must coordinate probability, adversarial games, invariants, assumptions and bounds, which can be provided by a machine-checked framework named EasyCrypt. Although all objects may appear in available context, LLMs still struggle because proof-theoretic dependencies are typically implicit in a linear representation and distributed across multiple programs. So, this paper presents Scratchy, a visual-scratchpad approach that exposes these dependencies for multimodal generation. Given the natural-language security description, with formal context and target propositions, the proof objects can be normalized into a typed proof-relation graph. Then a structure-preserving visual compiler transforms the graph into the formula-rich visual proof state that guides a multimodal model in generating the EasyCrypt proof. Also, the Scratchy-eval, a 114-task dataset derived from reliable official EasyCrypt files, has been introduced. It contains 64 security-form proof generations and 50 multiple-choice knowledge tests. After a series of evaluations, covering semantic grounding, relational invariants, and game reductions, classical LLMs like GPT-5.6-Sol and Claude-Opus-5 have gained a clear advantage from Scratchy's structured visual proof states. This contrast suggests that explicit proof structure can make the improvement and multimodal proof-state representation as a promising direction for computer-aided cryptography.
Read the full abstract

Analysis of PDEs

Determination of Wave Speed from point sources

We consider the wave equation with variable wave speed in Euclidean space, with point sources modeled by Dirac delta initial displacement data. There are three special subsets of the whole space: (1) the source set where the delta initial conditions are supported, (2) the unknown set where the wave speed is not known a priori, and (3) the receiver set where the waves are measured. The inverse problem is to reconstruct the wave speed uniquely in the unknown set from an unlabeled collection of waves generated by point sources and measured in the receiver set. We use propagation of singularities and sharp finite speed of propagation to reduce this data to geometric travel-time data, whose form depends on how the three sets lie in relation to each other. We give three scenarios where this procedure leads to unique determination of the wave speed.
Read the full abstract

All authors Joonas Ilmavirta, Teemu Saksala, Ashwin Tarikere, Gunther Uhlmann

High Energy Physics (Theory)

The Wilson-line-dressed charged sector of scalar QED: superselection and the infraparticle

We develop a manifestly gauge and Lorentz invariant perturbative calculus to study the finite time and distance evolution of the electrically charged states created by Wilson-line-dressed scalar field operators in scalar quantum electrodynamics. By an Osterwalder-Schrader reflection and Euclidean gluing, the overlap of two dressed states is expressed as a Euclidean 2 point function of dressed operators, which is then evaluated in renormalized perturbation theory with a Stueckelberg photon mass as a gauge, BRST and Lorentz invariant infrared cutoff. This renders explicit and computable in closed form several structural features of the charged sector that have so far been accessible mainly through non-perturbative or algebraic arguments. We find that states dressed by non-parallel Wilson lines are “cloud orthogonal". Their overlap vanishes as a power of the infrared regulator, with an exponent that depends only on the angle between the two dressings and is an infrared analogue of the cusp anomalous dimension with the difference that it is one-loop exact whenever the charged matter is mass-gapped. Charged states are thereby superselected by the orientation of their dressing. Within a superselection sector the 2 point function is infrared finite but exhibits the infraparticle. The mass shell pole is replaced by the edge of a branch cut. As a result, the asymptotic spread of a dressed charge is a power law in proper time with a one-loop-exact computable exponent depending on the dressing and on the direction of motion. We show that although cloud orthogonality and the infraparticle are each extremely sensitive to fundamental infrared cutoffs, the infraparticle scaling law persists across a wide window of proper time.
Read the full abstract

Analysis of PDEs

Concavity Properties of Robin Solutions on $C^{3,1}$ Uniformly Convex Domains

We prove that, on a bounded uniformly convex domain of class $C^{3,1}$, the first Robin eigenfunction is strictly log-concave and the Robin torsion function is strictly $1/2$-concave for all sufficiently large Robin parameters. This means that Conjecture 1.1 of Andrews, Clutterbuck and Hauer holds for uniformly convex $C^{3,1}$ domains in any dimension, and settles an open problem posed by Crasta and Fragalà when the domain has $C^{3,1}$ regularity. Our proof derives suitable uniform $C^2$-decay estimates by maximum principle arguments, thereby removing the imposed higher-order boundary regularity assumption required previously.
Read the full abstract

Information Retrieval

Relevance is not enough: A Communication-Oriented Retrieval System for Consequential Scientific Question Answering

AI systems increasingly answer scientific questions about health, safety, and the environment. But most retrieval-augmented generation systems are tuned to provide factually correct, on-topic answers rather than to help non-experts understand what those answers mean for their lives and decisions. We focus on consequential scientific questions whose results directly shape people's lives and study them through a public water-quality communication system, where residents and community leaders interpret the findings and choose actions. Their experiences show that on-topic answers can still be insufficient without explanation and context and that the emotional weight of risk information cannot be ignored. Our system first classifies each question by reasoning type (for example, causal versus policy-based), then generates follow-up questions to identify missing evidence and retrieve it. One component clearly distinguishes between what is known and what is uncertain, while another rewrites scientific details into accessible language, using persona-based styles, such as a caring neighbor or an administrative official, to adapt tone and readability. Ablations on over $160$ questions show that the system uses $\textit{an order of magnitude less context}$ and, in several configurations, improves human-rated completeness. A completeness metric co-designed with community members and a fine-tuned learned judge reveal that standard relevance scores explain about $1\%$ of variation in human completeness ratings, and even the tuned judge only moderately aligns with humans, indicating that completeness is a distinct human-centered objective that current metrics do not reliably capture.
Read the full abstract

All authors Avina Nakarmi, Naga Datha Saikiran Battula, Anthony Diaz, Aritra Dasgupta

Robotics

RefGuard: Identity-Aware Language-Guided Robot Manipulation via Joint Target-Anchor-Frame Grounding

Vision-language-action (VLA) models have substantially advanced language-guided robot manipulation, yet reliable execution still hinges on identifying which physical object an instruction refers to. In cluttered scenes containing repeated objects, ambiguous anchors, or frame-dependent spatial terms, a robot can execute a geometrically valid action on a semantically compatible but unintended instance; we call this failure an identity switch. The referent is jointly determined by three coupled latent variables: the target, the anchor, and the reference frame, so committing to any one of them before execution turns residual ambiguity into a silent and irreversible error. We propose RefGuard, an identity-aware grounding framework that delays commitment by maintaining a joint posterior over all three variables. RefGuard builds a frame-conditioned object-centric scene graph from RGB-D observations, separating frame-independent geometry from directional relations, and routes the posterior through a decision policy that executes, clarifies, reobserves, or aborts. On a real UF850 arm, RefGuard records no identity switch on any ambiguity-stress trial and executes correctly on 90.0% of them, whereas fine-tuned VLA and LLM (Large Language Model)-based baselines switch identity in 33-46% of the same trials, while retaining 93.3% success on unambiguous scenes and recovering from post-grounding scene changes in 86.7% of trials. On a 3200-episode procedural suite, it raises correct execution on solvable instructions from 56.6% to 80.5% over the ablation that commits to the anchor and frame before the target, while deferring less often (19.5% vs. 43.4%).
Read the full abstract

All authors Lan Wei, Kangyi Lu, Yongchen Wang, Chenmeng Bi, Qi Chen, Hanlin Niu, Yip Fun Yeung, Dandan Zhang

Group Theory

Gallai Decomposition of Ordered Groups: Subgroups, Quotients, and the $N$-free Case

We study Gallai decomposition for groups equipped with two-sided invariant partial orders. The key algebraic step extends to arbitrary binary relations compatible with the group operation: if all left and right translations preserve a binary relation $\rho$, then every least strong module $S_{\rho}(\e,g)$, $g\ne\e$, is a subgroup. For a partial order this subgroup is convex. Thus the robust modules through the identity of an ordered group form a canonical chain of convex subgroups, with each canonical factor $H/H^-$ prime, totally ordered, or equality-ordered. We characterize exactly the subgroups that are modules, show that they form a complete sublattice of the subgroup lattice, establish overlap and inheritance results for arbitrary subgroups, and prove compatibility with quotients by normal strong subgroups. For $N$-free ordered groups the prime factors disappear. Using the robust-module decomposition of cographs, we characterize all two-sided invariant $N$-free partial orders by reduced admissible two-coloured subgroup chains, with totally ordered and equality-ordered canonical factors; the order is determined by the first nontrivial factor of each element. We also determine how the canonical decomposition restricts to arbitrary subgroups, characterize finite width and prove width divisibility for subgroups, and show that every reduced two-coloured chain is realized by an $N$-free ordered abelian group whose canonical factors are isomorphic to $\mathbb Z$.
Read the full abstract

Information Retrieval

Closing the Long-Short View Gap in Sequential Recommendation without Cached History

Sequential recommenders are typically trained on long user histories to capture rich behavioral signals, yet serving with training-length sequences is often impractical due to real-time efficiency constraints. Directly using only recent behaviors leads to a severe performance drop. To bridge this gap, existing approaches compress user histories into persistent per-user states, storing and retrieving them at inference time; while effective, they impose non-trivial infrastructure overhead and offer little remedy in cold-start scenarios. In this paper, we empirically identify two structural flaws rooted in geometric properties and dataset sparsity, and propose a novel two-stage framework to close the long-short-view performance gap. Specifically, in the first stage, we replace the commonly used dot-product with angular similarity scoring and leverage a modified softmax to counter prefix position bias. In the second stage, we fine-tune only bias and LayerNorm components, which are universal to standard sequential backbones, for further improvement. Both stages are guided by carefully designed learning objectives. Extensive experiments on two representative backbones across three public datasets demonstrate the effectiveness of our proposed framework.
Read the full abstract

All authors Lingfeng Shi, Chengkai Huang, Lina Yao, James Caverlee

Computer Vision

Branch-Centric Tokenization and Test-Time Augmentation for Skeleton Generation

Automatic skeleton generation involves predicting both joint positions and skeletal connectivity. However, existing approaches struggle to encode branch structures into token sequences and do not use test-time computation effectively. We study these choices within a unified autoregressive framework. First, we introduce branch-centric tokenization, a branch-aware representation that places structurally related elements next to each other and encodes connectivity directly in the sequence. Compared with standard BFS-style serialization, this representation yields more compact sequences. Second, we introduce view-augmented generation, a test-time augmentation procedure that applies axis-aligned rotations to the input mesh, maps all predictions back to a common frame, and selects the final skeleton based on mesh coverage and consistency among predictions from different views. Experiments show that our method achieves better skeleton prediction accuracy than state-of-the-art methods. In particular, our method reduces the CD-J2B error by 16.9% on the Articulation-XL2.0 dataset compared to the strongest directly comparable baseline, Auto-Connect. Qualitative results on in-the-wild meshes further demonstrate generalization across diverse inputs.
Read the full abstract

All authors Zhengyuan Li, Chuanyu Pan, Yuanming Hu, Raymond A. Yeh

Exactly Solvable and Integrable Systems

On Bäcklund transformations preserving the Darboux integrability of hyperbolic equations

This paper deals with two kinds of Bäcklund transformations for scalar hyperbolic partial differential equations. We prove that both these types of transformations map solutions of a Darboux integrable equation into solutions of, generally speaking, another but also Darboux integrable equation. The latter fact can be used to roughly check the completeness of a list of Darboux integrable equations. To illustrate this, we apply the above transformations to several equations from a well-known list of Darboux integrable equations and, as a result, obtain a Darboux integrable equation which is absent in this list, but is already known at present. As a generalization of the last equation, we construct a family of Darboux integrable equations that is parametrized by three arbitrary functions, each of which depends on two arguments. This family is probably new.
Read the full abstract

Nuclear Theory

Nucleon Correlations in Unstable Nuclei via Knockout Reactions

We review the physical ideas and reaction theory underlying one- and two-nucleon removal from fast projectiles, quasi-free proton-induced reactions such as $(p,2p)$ and $(p,pn)$, and the use of exclusive momentum and coincidence observables to diagnose correlations. Particular emphasis is placed on the distinction between independent-particle occupancy, long-range collective and pairing correlations, tensor-driven neutron-proton correlations, and short-range correlations generated by the repulsive core and noncentral components of the nuclear interaction. The discussion develops the overlap-function language, spectroscopic factors, eikonal stripping and diffraction, distorted-wave impulse approximation, factorization and spectral functions, center-of-mass and recoil effects, two-nucleon amplitudes, pair densities, and the relation between measured cross sections and ab initio or shell-model structure. The well-known reduction of experimental single-particle strength relative to simple shell-model expectations, its dependence on separation-energy asymmetry, and the continuing debate over reaction-model systematics are treated in detail.
Read the full abstract

Mesoscale and Nanoscale Physics

Purely Electric-Field Control of Topological Magnetism in Two-Dimensional Magnets

Electrical control of topological magnetism is central to realizing energy-efficient topological spintronics. Yet most electric-field approaches modify the competing magnetic interactions through a nonselective rearrangement of low-energy electronic states, yielding coarse magnetic phase control that often requires external magnetic fields to stabilize topological quasiparticles, while few schemes solely based on electric fields are restricted to specific conducting materials. Here, we establish a general approach for purely electric-field control of topological magnetism, in which an applied electric field electrostatically dopes a selected orbital-angular-momentum-polarized band edge of a two-dimensional (2D) van der Waals (vdW) magnetic semiconductor via proximity to an adjacent nonmagnetic vdW metal. We show that the resulting electrostatic doping predominantly tunes magnetic anisotropy of the 2D magnet, while leaving exchange interaction and Dzyaloshinskii-Moriya interaction nearly unchanged, thereby reversibly driving the system through ferromagnetic, skyrmion, spiral, and bimeron phases. We demonstrate this mechanism for CrBr3/graphene and Cr2Ge2Te6/TaS2 vdW heterostructures hosting electron and hole pockets of different orbital characters. These results establish a broadly applicable strategy for purely electric-field control in topological spintronics.
Read the full abstract

All authors Zhonglin He, Naafis Ahnaf Shahed, Kai Huang, Himanshu Mavani, Evgeny Y. Tsymbal

Numerical Analysis

Fourier-cosine Tensor Trains for Density Recovery and Expectation Calculation

The Fourier-cosine (COS) method of Fang and Oosterlee (2008) recovers densities and computes expectations semi-analytically from characteristic functions (ch.f.s). In higher dimensions, both expansion synthesis and COS coefficient-tensor construction scale exponentially with dimension. We address this curse of dimensionality with two tensor-train (TT) methods. A straightforward TT decomposition of the COS coefficient tensor gives COS-TT, which removes the exponential cost of online synthesis but not of offline decomposition. Our main contribution, COS-TT-CHF, instead compresses a ch.f. sample tensor and maps it to the original Fourier–cosine coefficient tensor through an alternative COS representation. Apart from the model-dependent cost of one ch.f. evaluation, both offline and online costs scale linearly with dimension, while error grows only algebraically. COS-TT-CHF introduces two additional errors: frequency-domain truncation and discretization errors. Both are controlled by parameter-selection rules from theoretical error analysis. We further propose an approach for expectations of nonseparable functions of additive scalar aggregates by computing the ch.f.s of the aggregates using tensorized COS methods, reducing the original multidimensional problem to a one-dimensional COS calculation. Closed-form formulas are derived for the TT-contraction integrals arising when the aggregate is a weighted sum, as in European basket option pricing. Experiments under geometric Brownian motion (GBM) and variance gamma (VG) show that, with the parameter-selection rules, COS-TT-CHF remains accurate through 150 dimensions for GBM and 100 for VG within a 30-minute offline computation budget on a laptop. As a by-product, our Fourier-cosine TT constructions yield cosine-basis functional TTs (FTTs) under weaker assumptions than in spectral FTT literature.
Read the full abstract

Software Engineering

Beyond Lexical Metrics: Sentence-Embedding Detection of Reviewer Habituation in AI Code Review

Code review is a key quality checkpoint between AI-generated code and production. As AI coding agents submit pull requests at scale, it is unclear whether reviewers reduce scrutiny with repeated exposure and whether review comments reveal this change. We study 11,429 reviews from 400 repeat reviewers over 207 days, paired with 10,104 human-authored inline comments from AIDev. Approval rates rise from 30.5% in reviewers' early periods to 36.6% in late periods (Wilcoxon p = 8.6 x 10^-8; Cohen's d = 0.25). However, four hand-crafted linguistic features - lexical diversity, Shannon entropy, technical specificity, and constructive actionability - show no monotonic decline across exposure deciles (all Spearman absolute rho <= 0.53, p >= 0.11; Bonferroni-corrected Mann-Whitney p >= 0.36). A logistic-regression classifier based on these features reaches F1 = 0.485, below the majority-class baseline. Sentence-embedding structure does carry signal: reviewers' late-period comment centroids shift farther from their early-period centroids than under within-reviewer random permutations (Wilcoxon p < 0.001), and a small MLP using three embedding statistics reaches F1 = 0.74 under 5-fold reviewer-stratified cross-validation. Granger analysis shows that approval-rate changes predict later shifts in technical specificity at all tested lags (p < 0.001), while the reverse direction is significant at only one of four lags. Reviewer adaptation is therefore detectable in latent distributional structure rather than in classical lexical metrics, and language shifts follow rather than precede changes in approval behavior.
Read the full abstract

Computation and Language

SLATE: Are AI-Generated Slides Educationally Effective? A Benchmark for Language Teaching Quality and Learner Knowledge Acquisition

LLMs have achieved remarkable capabilities in generating language teaching slides. However, a critical mismatch persists between visual polish and actual instructional effectiveness. To address this gap, we introduce SLATE (Slide-based Learning Assessment for Teaching Effectiveness), the first benchmark that evaluates AI-generated language teaching slides through instructional effectiveness and learner knowledge acquisition. SLATE transforms linguistics olympiad puzzles from low-resource languages with negligible web presence into 90 standardized instructional units comprising 1,133 assessable items, paired with a structured course outline and matched near- and far-transfer test sets. This pretest-posttest design eliminates pretrained knowledge leakage, ensuring gains reflect learning rather than prior recall. Using VLMs as scalable learner proxies and directionally supported by a three-system human pilot, our results show that content validity exhibits a weak association with learning gain, while pedagogical design exhibits a robust positive association. Moreover, most systems show a significant gap between near- and far-transfer accuracy, and even frontier models can produce negative learning gains. SLATE reveals a dissociation between artifact quality and instructional effectiveness, calling for a paradigm shift in how generative teaching systems are built, evaluated, and deployed.
Read the full abstract

All authors Jingzhuo Wu, Jiajun Zhang, Liu Yi, Leqi Zheng, Yuheng Jing, Xinyuan Zhou, Quan yang

General Relativity and Quantum Cosmology

Alternative neural-network follow-up for all-sky FrequencyHough in continuous gravitational-wave searches

Continuous gravitational waves are long-lived signals emitted by spinning neutron stars (NSs) or boson clouds around black holes. While the Milky Way is expected to host $\mathrm{O}(10^{8}-10^{9})$ NSs, only $\mathrm{O}(10^{3})$ are currently identified through electromagnetic observations. This large observational gap motivates searches based on alternative messengers. In particular, all-sky searches for continuous gravitational waves (GWs) offer a unique opportunity to detect NSs that are electromagnetically silent or otherwise undiscovered. By exploring a broad region of parameter space without any a priori source information, these searches can probe the vast hidden NS population of our Galaxy. In this work, we present a novel way to include a Neural Network (NN) classifier into an all-sky search strategy for isolated NSs. Starting from candidates produced by the FrequencyHough pipeline running on data from ground-based detectors such as LIGO and Virgo, the proposed method improves sensitivity without increasing computational cost and can be parallelized across multiple detectors to enhance detection probability and reduce false alarms. We analyze data from the third observing run (O3) — April 1, 2019 to March 27, 2020 — and focus on the frequency range [129, 229] Hz and spin-down interval [$-2.5\cdot10^{-9}, 1.0\cdot10^{-9}$] Hz/s. The model is trained on noise constructed to be statistically consistent with real O3 data and tested on real O3 data. Results show robust performance in distinguishing signal from noise, even at low strain, and successful identification of hardware injections outside the training spin-down range.
Read the full abstract

All authors Martina Di Cesare, Pia Astone, Rosario De Rosa, David Keitel, Cristiano Palomba, Marco Serra

Formal Languages and Automata

From Ramsey-Based to Congruence-Based Constructions for Büchi Complementation

The very first construction by J. Richard Büchi himself for complementing a Büchi automaton relies on a fundamental lemma about the division of an arbitrary infinite word into consecutive finite words, which was cleanly proven by invoking a specialized theorem of Ramsey. For that reason, constructions of similar nature have subsequently been labeled as Ramsey-based. Nevertheless, it suffices to have a weaker form of the lemma where the finite words come from a finite number of congruence classes, rather than arbitrary classes, that form a partition of the set of all finite words. The weaker lemma, with support of nicer properties from a congruence, can be proven without Ramsey's theorem. A commonly adopted improvement on such complementation constructions also requires the working of a congruence. This paper recounts the history and reviews using more contemporary terminology wherever possible the relevant concepts and results, to advocate renaming of Ramsey-based constructions as congruence-based constructions.
Read the full abstract

Graphics

RBF Your SDF: Radial Basis Function Interpolation of Signed Distance Fields with Implied Tangent Points

Signed distance fields (SDFs) are a popular implicit representation of geometry. Converting a discrete set of SDF samples into an explicit surface is a fundamental problem in geometry processing. Traditional reconstruction methods such as marching cubes and dual contouring ignore the geometric information carried by samples far from the surface. Recently, Sellán et al.[2023] and several follow-up works leveraged the tangent-sphere structure of SDFs; every sample implies a point on a sphere tangent to the surface. However, these approaches extract the zero-level set via surface reconstruction, which considers only points and normals on the surface and ignores the remaining samples. We propose an approach that marries the tangent-sphere observation with radial basis function interpolation of all data, the implied surface points and the original data. By detecting spheres with extremely constrained tangent points, a configuration geometrically forced at sharp surface features, we identify and preserve surface corners that surface reconstruction-based methods systematically round. A partition-of-unity decomposition allows our method to scale efficiently to large grid resolutions. Our reconstructions improve both Chamfer and Hausdorff accuracy at every tested resolution.
Read the full abstract