Materials Science
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Wednesday 2 September
Altermagnetic materials exhibit spin splitting in their electronic band structures while maintaining zero net magnetization. However, conventional high-symmetry k-paths generally hide this splitting because they follow symmetry lines that in most cases enforce spin degeneracy. We present AlterSeeK-Path, an open-source Python tool that systematically constructs generalized band-structure paths for collinear altermagnets. The method selects the centroid of the conventional irreducible wedge used in routine band-structure calculations as the representative general k-point, maps it to a spin-flip-related partner, and inserts paired segments through these points into the standard high-symmetry path; these segments systematically sample the interior of the irreducible wedge. We demonstrate the construction for all 54 three-dimensional combinations of extended Bravais lattice type and spin Laue group across the six crystal systems that support collinear altermagnetism, and for the 12 two-dimensional cases spanning the four compatible two-dimensional Bravais lattices. Representative band structures are shown for the distinct lattice/path cases. With AlterSeeK-Path, these band structures can be constructed with essentially the same effort as conventional band structures.
Analysis of PDEs
Divergence-Free Approximation in Sobolev and Lebesgue Spaces on General Unbounded Domains with Applications to Energy Equality in Fluid Dynamics
We construct divergence-free, vector-valued approximation functions on the whole space $\mathbb{R}^d$, $d\geq 2$, as well as on general unbounded domains of uniform $\mathrm{C}^{1,1}$-type. These approximations converge simultaneously in both Sobolev and Lebesgue spaces. In the whole-space setting, we employ Bogovskiĭ\ operators to construct such approximations, thereby extending the approximation theory developed for smooth bounded domains and the simultaneous approximation framework introduced by \emph{Fefferman, Hajduk, and Robinson, {Proc. Lond. Math. Soc.} (3) \ {125} (2022), no.~4, 759-777}. As an application, we establish energy equality for Leray-Hopf weak solutions of the incompressible convective Brinkman-Forchheimer (CBF) equations on $\mathbb{R}^d$, $d\in\{2,3\}$ covering both the critical and supercritical regimes. For general unbounded domains, we employ the resolvent operator associated with the Stokes operator, developed by \emph{Farwig, Kozono and Sohr, {Acta Math.}, {195} (2005), 21-53}, to obtain simultaneous approximation results. We also establish a generalized version of the classical Lions-Magenes lemma, which is of independent interest. Finally, by combining this result with the simultaneous approximation framework for unbounded domains, we establish energy equality for weak solutions of the CBF equations on general unbounded domains.
Read the full abstract
Mesoscale and Nanoscale Physics
Emergence of the magnetic octupole Rashba-Edelstein effect from spin-orbit entanglement
Magnetic multipoles have attracted growing interest as order parameters and dynamical degrees of freedom in unconventional magnets, yet it remains unclear how broadly they can emerge as active electronic degrees of freedom. Here, we show that spin-orbit-entangled multiorbital states can host magnetic octupole (MO) degrees of freedom. In Rashba systems, this gives rise to an MO Rashba texture accompanying the J-Rashba texture. Remarkably, spin-orbital entanglement can realize a pure-MO limit in which the spin and orbital-angular-momentum textures vanish while the MO texture remains finite. An applied electric field converts this texture into a nonequilibrium MO polarization through an MO Rashba-Edelstein effect. Around the pure-MO regime, the MO Edelstein response dominates over the conventional spin response, showing that multipolar responses need not be small corrections to dipolar spin physics. Our results establish an interface-based route for electrically generating nonequilibrium multipolar magnetic polarization in spin-orbit-coupled systems.
Read the full abstract
Materials Science
Systematic display of the spin splitting in band structures of representative altermagnetic crystals
In this work we demonstrate a novel approach to exhibit the unique spin splitting that is typical of collinear altermagnets. This approach is to plot band structures on Brillouin zone paths that sample general k-points to show a representative picture that corresponds to Brillouin zone averages. This is in contrast to conventional band structure plotting which plots band structures on the highest-symmetry points and lines, and thus in many cases can fail to show any altermagnetic spin splitting at all. Our investigation compares the new approach with the band structures of collinear altermagnets using conventional band structure plotting. We report the band structure and symmetry analysis for MnTe, CrSb, SmFeO3, ScCrO3, LaMnO3, TlCrO3, HoFeO3, InCrO3 and DyFeO3. This result clearly demonstrates the advantage of this novel method for displaying the spin splitting of collinear altermagnets.
Read the full abstract
Machine Learning
Do Tabular Foundation Models Know Physics? Contamination, Units, and the Deterministic Limit
Tabular foundation models (TFMs) learn to fill in tables the way language models fill in text — and tables are arguably the format in which most physical measurement arrives. Did they learn any physics in the process? They are Bayesian by construction, so the question is what their prior contains. We probe it directly, evaluating four of them ($\texttt{TabPFN-3}$, $\texttt{TabICLv2}$, $\texttt{TabDPT}$ and $\texttt{Real-TabPFN-2.5}$) against six baselines on datasets sampled from 316 physical equations, in and out of domain. TFMs dominate, out of the box and after tuning. But they struggle to represent noiseless deterministic mechanisms in their uncertainty, and show no sign of exploiting physical units: they interpolate physics without yet being able to act as physical models.
Read the full abstract
Representation Theory
Quantum wreath products and Yang-Baxter equations
We extend the setup of Lai-Nakano-Xiang's quantum wreath product algebras to symmetric monoidal categories and deformed braid relations. Under some mild assumptions, we drastically simplify the criterion for such algebras to have PBW basis; curiously, the new criterion is phrased in terms of associative and quantum Yang-Baxter equations. As an application, we clean up some basis results in the existing literature, and propose a certain generalization of quiver Hecke algebras.
Read the full abstract
Data Structures and Algorithms
Finding a Shortest Vector and More in $2^{n/2+o(n)}$ Time using $q$-ary Coset Difference Tree
This paper presents a new randomized algorithm for solving the exact shortest vector problem. For the $n$-dimensional lattice $\mathcal L$, our algorithm runs in time and space $2^{n/2+o(n)}$. Our algorithm can be viewed as a $q$-ary analogue of the midpoint Hessian for an odd prime $q$; more precisely, we use the fact that, for a shortest vector $v$, the gradient (rather than Hessian) of the periodic Gaussian function at $v/q$ is nearly proportional to $v$ (up to sign), even after aggregation over a relatively large random affine coset. We compute the relevant coset gradient along a chain of intermediate lattices using a combinatorial procedure inspired by Wagner's generalized birthday algorithm, yielding the $2^{n/2+o(n)}$ time and space complexity. A variant of the algorithm solves the exact closest vector problem on every input $(y,\mathcal L)$ with a distance guarantee $\operatorname{dist}(y,\mathcal L)\le 1.039\lambda_1(\mathcal L)$ within the same time and space complexity. This guarantee holds for a random target and a random lattice drawn according to the Haar-Siegel measure. Thus, this algorithm solves a closest vector problem on such random instances in time and space $2^{n/2+o(n)}$.
Read the full abstract
Combinatorics
Bounded Twin-Width Tournaments are $\dchi$-Bounded
For a tournament $T$ and a vertex ordering $\prec$, let $T^{\prec}$ be the graph of backward arcs in $\prec$. The diclique number of a tournament is $\domega(T)=\min_{\prec}\omega(T^{\prec})$, and the dichromatic number is $\dchi(T)=\min_{\prec}\chi(T^{\prec})$. We prove a mixed parameter transfer theorem: for all $k$ and $r$, if $\tww(T)\le k$ and $\omega(T^{\prec})\le r$, then the ordered twin-width of $(T^{\prec},\prec)$ is bounded by a function of $k$ and $r$. The proof combines the regular-semigrid theorem for ordered graphs with permutation-encoding obstructions to bounded twin-width in tournaments. Together with polynomial $\chi$-boundedness of graphs of bounded twin-width, this implies that tournaments of bounded twin-width are $\dchi$-bounded by $\domega$, resolving a conjecture of Aboulker, Aubian, Charbit, and Lopes.
Read the full abstract
Soft Condensed Matter
Relating solute interactions to interfacial properties
Liquid interfaces are both ubiquitous and also critically important for many commercial products, modern technologies, biological processes, and environmental phenomena. The properties of these interfaces can depend quite sensitively upon the composition of the bulk liquid phase. In this work, we develop a dilute solution theory (DST) for the influence of dilute cosolutes upon the interface between coexisting liquid and vapor phases. We employ a grand canonical perturbation theory to rigorously relate the interfacial free energy to the concentration of the liquid solution. We leverage a corresponding Gibbs ensemble to treat liquid-vapor coexistence and to eliminate the contribution of the bulk phases from this free energy. We express the coefficients of the resulting expansion in terms of microscopic partition functions. By treating solute-solute interactions to lowest order, we distinguish between the intrinsic and effective interfacial preferences of solutes. While the former reflects only solute-solvent interactions, the latter depends upon the solution composition and reflects the influence of solute-solute interactions. We assess this DST with molecular dynamics simulations of binary and ternary mixtures of Lennard-Jones spheres. These simulations demonstrate that DST accurately models the interfacial properties of these systems up to relatively high concentrations. Moreover, the simulations illustrate the impact of attractive solute-solute interactions in converting weak intrinsic surfactants into weak effective depletants.
Read the full abstract
High Energy Physics (Phenomenology)
Next-to-leading order QCD corrections to fully charm tetraquark hadronic decay
We compute the next-to-leading order (NLO) QCD corrections to the light hadron decays of fully charm tetraquarks within the nonrelativistic QCD (NRQCD) factorization framework. The short-distance coefficients for the $gg$ and $q\bar{q}$ final states from fully charm tetraquarks are obtained analytically. The NLO corrections are found to be significant, altering the LO predictions by about $170\%$ for the $0^{++}$ state and $30\%$ for the $2^{++}$ state. The resulting $\mathcal{R}_{\mathrm{LH}}$ values are of order $10^{-5}$ to $10^{-4}$ MeV, about three orders of magnitude larger than the diphoton channel, as expected for strong-interaction decays. Our results provide updated theoretical predictions for future experimental studies of fully charm tetraquark decays.
Read the full abstract
Artificial Intelligence
Measurement-Driven Sub-Network Selection for On-Premise Retrieval-Augmented Factory Agents
On-premise assistants can give factory workers conversational access to machine documentation, but models capable of the task rarely fit shop-floor hardware. We show that after structural compression and retrieval-grounded adaptation, model size is no longer a reliable predictor of adapted answer quality: general capability falls almost linearly with parameter count, while judged retrieval-augmented answer quality does not. We therefore treat deployment as a post-adaptation selection problem, committing one sub-network per device on judged answer quality and measured on-device throughput under a configurable general-capability floor and memory budget; rules that optimize size, speed, or quality alone each give up capability or throughput. A weight-shared supernetwork trained with sandwich-style in-place distillation keeps this selection inexpensive. In a manufacturing-manual case study, extraction costs 13.7 percent of the unpruned model's judged quality and retrieval-grounded distillation returns it to within 4.6 percent, recovering two thirds of the loss, and the same assistant runs across three heterogeneous edge tiers at 1.3 to 5 watts standby.
Read the full abstract
Other Condensed Matter
Magnetoelectric Phase Transition and Axion Dynamics
Magnetoelectric phase transitions have been experimentally studied, but no macroscopic theory has been proposed to explain their dynamical origin. In this work, we assume that the axion quasiparticle with frequency undergoes a condensation like process. We show that these magnetoelectric phase transitions can be described within a Ginzbur Landau framework by introducing a coupled dynamic parameter, the axion angle, which is proportional to the magnetoelectric coeffcient. We derive relations between the static axion angle, the axion frequency, and the phase transition temperature for different magnetoelectric materials, respectively, and compare these calculations with existing experimental results. We also connect the artificially designed Dzyaloshinskii Moriya interaction with the axion condensate like process, so that the relation between the static axion angle and the experimentally measured frequency shift can be obtained.
Read the full abstract
High Energy Physics (Phenomenology)
Next-to-Leading-Order Multi-Jet Merging of Higgs Boson Production via Vector Boson Fusion in the Presence of Anomalous Couplings
We present the results of a detailed simulation study of Higgs boson production via vector boson fusion (VBF) at the Large Hadron Collider (LHC) in the presence of anomalous couplings. The analysis utilizes the Herwig event generator interfaced with VBFNLO, employing a multi-jet merging scheme that combines NLO accurate matrix elements for Higgs plus 2- and 3-jet multiplicities with LO accurate matrix elements for Higgs plus 4-jet multiplicities. We investigate both matching and merging setups to consistently combine these hard scattering processes with QCD parton showers. Finally, we demonstrate the robustness of azimuthal angle observables against QCD radiative corrections across these frameworks.
Read the full abstract
Cosmology and Nongalactic Astrophysics
Gravitational Imprints of Axion Clouds on Black-Hole Orbital Dynamics
Black holes of primordial or astrophysical origin immersed in axion dark matter provide a natural laboratory for exploring the interplay between compact objects and ultralight scalar fields. While previous studies have primarily focused on superradiance and horizon-scale phenomena, the gravitational response of the surrounding axion distribution has received comparatively little attention. We investigate the dynamics of particles orbiting Schwarzschild and Kerr primordial black holes embedded in scalar-field dark matter halos. Using first-order perturbation theory, we show that the ambient halo redistributes the gravitationally bound axion cloud, shifting the characteristic radius of the gravitational atom. Numerical integrations of timelike geodesics demonstrate that the maximum orbital response consistently occurs at this perturbed radius, establishing a direct correspondence between the microscopic structure of the axion wavefunction and the macroscopic dynamics of orbiting particles. Our results identify a new purely gravitational signature of scalar-field dark matter around primordial black holes and provide a framework for probing mixed axion–primordial black hole dark matter scenarios.
Read the full abstract
Statistical Methodology
Off-policy causal estimation in networks
In the presence of interference, where the treatment assigned to one unit can affect the outcomes of others, many causal estimands depend on the treatment-assignment policy under which the experiment is conducted. This policy dependence creates a fundamental challenge for off-policy estimation, where the goal is to estimate causal quantities under a hypothetical intervention policy different from the one used to collect data. We study this problem of off-policy estimation of causal effects for heterogeneous Bernoulli policies. By representing exposure-weighted potential outcomes in the biased Fourier basis of the experimental design, we construct, for any prespecified Fourier subspace encoding the assumed interference structure, the unique minimum-$L^2$ weight that transports every function in that subspace. Global and local inverse-probability weights, linear-interference weights, and no-interference weights are special cases. The weight variance is a structured chi-square distance between the experiment and target policies. When the assumed interference structure is misspecified, the introduced bias couples the omitted outcome spectrum with the corresponding policy-shift coefficients, yielding a sharp robustness bound and a bias-variance trade-off. A Fourier-neighborhood-overlap condition gives consistency under structured interference, and we state a Doob-martingale central limit theorem for off-policy estimators. As the variance is not identified, we derive identifiable bounds and associated conservative estimators of the variance. Simulations illustrate these theoretical results for the design and analysis of experiments under network interference and design mismatch.
Read the full abstract
Probability
The Sharp Rate of Probabilistically Strong Convergence to the KPZ Equation
One of the most common descriptions of solutions of singular SPDEs is their characterisation as the limit of solutions of renormalised smooth random PDEs. We quantify the speed of this convergence in the case of the KPZ equation. In particular, we show that the naïve guess that the rate is given by the distance of the noise regularity from the endpoint regularity for well-posedness is not correct. Instead, we obtain convergence at the larger rate $1/2$ and show that this rate is sharp. This is achieved by considering the equation satisfied by the rescaled error, which is critical for variance blowup, and showing that this equation has a limit given by an affine linear singular SPDE driven by a new, independent noise. This result can be alternatively interpreted as identifying the asymptotic size and law of fluctuations of the solutions of the KPZ equation driven by mollified noise around their singular limit.
Read the full abstract
Computation and Language
Untangling the Mechanisms of Misleading Context in Medical Question Answering
Large language models now answer medical questions with expert-level performance. However, the context these systems act on can be misleading, and misleading context can corrupt a model's medical judgment. To understand how misleading context corrupts this judgment, we examine the model's susceptibility to the context, disclosure of it, mechanism of corrupted reasoning, and monitorability of the decision. On the medical reasoning subset of MedMisBench, a clinician-reviewed question-answering benchmark of 8,627 questions, we inject two types of misleading context cues, fabricated evidence and a bare assertion. We test three reasoning models, two that expose their full reasoning trace and one frontier model that exposes only its response. All three are more susceptible to the assertion than to the fabricated evidence, adopting the asserted answer 10 to 27 points more often. The misleading cues are disclosed in 81 to 98% of traces but only 7 to 90% of responses, and the assertion is disclosed less often than evidence based cues. Resampling from reasoning traces without disclosure shows the two cues corrupt reasoning differently, evidence entering early and accumulating while the assertion redirects the conclusion near its end. An LLM monitor catches 78% of corrupted decisions at 5% false positives when reading an open model's trace with guidance, against at most 32% from any response. The misleading context that models are most susceptible to is disclosed least, and was caught reliably only from an open reasoning trace, which frontier providers withhold.
Read the full abstract
Software Engineering
The Import Tax: A Longitudinal Measurement of Startup Cost in the Python Ecosystem
Python programs pay for their imports at every process start, a cost that is invisible in steady-state benchmarks but dominant for command-line tools, test workers, and serverless cold starts. Python 3.15 adds explicit lazy imports (PEP 810) on largely anecdotal evidence. We measure the import cost of the 500 most-downloaded PyPI packages, every quarterly release since mid-2021, under six CPython versions (3.9-3.14) on two platforms (Apple M5/macOS and Intel Xeon/Linux), 66,816 cells in all, with targeted experiments that isolate individual effects. Values are for macOS (Linux in parentheses). Import cost is heavily skewed (median 12.3 ms, 99th percentile 400 ms). Among packages that import in more than 2 ms, namespace distributions excluded, the median package spends 68% of its import time in the standard library (68% on Linux with the interpreter on local storage, 79% on network storage). Precompiling bytecode removes almost all of the cold-start penalty on Linux and 92% of it for the median package on macOS, but only a third of the summed macOS penalty: packages with native libraries pay a security assessment of each library that the system has not seen before, again in every fresh environment. A package's own import code grows by a median 4.0% (5.0%) per year; installing each release with the dependencies of its time changes its trend by a median of only 0.15 (0.35) percentage points (CPython 3.9). CPython 3.14 runs package code faster than 3.9 but imports more and slower standard-library modules, a net 1.16x (1.07x) slowdown. Submodules that dependents import beyond the top level add nothing for the median package and at most 11.7 (49.5) ms for nine packages in ten. On CPython 3.15, global lazy mode cuts the imports that dependents perform to 0.41x (0.39x) of their cost, and 38 of 461 packages fail under it, many only on first use of a name. Harness, code, and dataset are public.
Read the full abstract
Lattice Field Theory
Double Parton Distributions in the nucleon: Lattice parameter dependence
To make full use of the HL-LHC, especially in the search for BSM physics, all Standard Model backgrounds have to be understood with increased precision. This is especially difficult for Double Parton Interactions (DPIs) and more generally Multi Parton Interactions (MPIs), which are very difficult to isolate precisely in experiment. DPIs are parameterized by Double Parton Distributions (DPDs) and the lattice determination of the latter is, therefore, especially attractive. In our previous work on DPDs on the lattice, we focused on the results obtained from one gauge ensemble with higher than physical pion mass. In this work, we extend our efforts by investigating artifacts introduced by finite volume and the dependence on the lattice spacing. Moreover, we consider the dependence on the mass parameters. This study is carried out by employing the Wilson-Clover fermion gauge ensembles generated by the CLS collaboration.
Read the full abstract
Computer Vision
Multi-Tool Image Editing Attribution in Facial Forgery
As generative AI tools become increasingly powerful and easy to use, people can easily edit portrait images with a prompt, necessitating the task of image editing attribution, which predicts the involved editing tools from the given image. Existing attribution methods hold the single-tool assumption and can only attribute a specific editing tool, but struggle to handle the more complex and increasingly common multi-tool editing scenarios, where artifacts left by different editing tools are composite and overlapped. To address this gap, we explore Multi-Tool Image Editing Attribution (MIEA), which aims to identify multiple editing tools involved in a multi-tool edited facial image. To simulate the real-life editing operations on facial images, we then construct a new dataset, MultiEdit, which contains 500k+ edited facial images and covers six types of editing tools that support face swapping (Deepfake) and various facial enhancements. Inspired by the findings from data analysis, we design DPEC, a multi-tool attribution method that can capture distinguishable, locality-aware editing tool traces from both spatial and frequency domains with the support of an error-based curriculum learning strategy. Experiments show \Method\ outperforms nine methods for facial images edited in at most five steps.
Read the full abstract
Artificial Intelligence
Bilevel Coordinated Reflection: A Game-Theoretic Approach to Multi-Agent LLM Systems
Multi-agent LLM systems commonly use an orchestrator to decompose a task for a team of workers and then improve through textual reflection. Despite strong empirical results, these systems lack a unified account of coordination, memory improvement, and the role of external verification. We model orchestrator-worker interaction as a bilevel coordination game: under bounded coupling, the workers' local-update game is an approximate potential game whose equilibrium slack is controlled by decomposition quality. We then analyse reflection as stochastic movement over semantic memory states. For free-form reflection, we derive a finite-time upper bound, prove worst-case tightness, and give a positive lower bound under a falsifiable persistent-harm condition. We further prove an information-theoretic impossibility result: no gate that observes only the generated transcript can improve uniformly over text-indistinguishable environments, whereas an environment-grounded gate can. Motivated by this separation, we introduce Stochastic Reflective Memory Ascent (SRMA), which accepts a candidate memory only after a grounded evaluation risk strictly decreases. Under calibration and non-degenerate corrective mass, SRMA converges exactly, geometrically or polynomially; matching constructions show that both rate regimes are order-tight. We also provide confidence gating for stochastic evaluation and re-anchoring guarantees for piecewise-stationary environments. Experiments instantiate these objects with environment-grounded metrics and test the predicted coordination and drift laws. On 500 SWE-bench instances, the complete Kimi-based system resolves 72.2% versus a 70.8% public mini-SWE-agent reference. Code: https://github.com/YihangChen9/Bilevel-Coordinated-Reflection
Read the full abstract
Artificial Intelligence
Repo-To-Skill: Distilling GitHub Repositories Into AI4AI Skills
Autonomous agents are beginning to carry out machine-learning (ML) research end to end. These agents combine a model backbone with a harness for planning, execution, memory, and verification, but this architecture still leaves domain-specific know-how outside the agent. We call this missing layer operational knowledge, the know-how that separates knowing a method from making it work. That knowledge is not absent from the field. It appears in repositories and papers, but in forms written for human readers and too large to load during a task. Once distilled into compact, verified skills, this knowledge can be reused across tasks rather than rediscovered during each run. We present DisCo, a skill-powered research agent that creates skills and uses them during research. Its distillation runs in two complementary forms: task-agnostic, condensing the field's widely used repositories into reusable skills, and task-oriented, producing the skills a concrete task calls for. The former, applied across the open ecosystem, yields the AREX-Skill Library, with 5,000+ verified skills distilled from 1,000 widely used ML repositories and organized into 20 areas and 178 capability families. With the GPT-5.5 backbone, research harness, and downstream execution budget held fixed, the skill-equipped research agent scores 134.3% higher on MLE-bench, 34.4% higher on PaperBench, 9.2% higher on FrontierCS, and 14.0% higher on PassNet than the same agent without skills. These gains come from adding distilled operating context under that fixed setup.
Read the full abstract
Computer Vision
Balancing Frequencies and Pixels in Flow Matching
Natural images follow a $1/f^2$ spectral distribution: most signal energy lies in the low spatial frequencies, while the perceptually important structures such as textures and edges occupy sparse high-frequency bands. Pixel-space reconstruction objectives, however, treat all spatial errors uniformly, causing low frequencies to dominate the optimization signal and delaying the learning of fine-scale details. In this work, we identify this objective-level spectral imbalance as a key inefficiency in training pixel-space flow models. To address it, we propose a Focal Log-Frequency Loss (f-loss), a spectrally balanced objective that equalizes the learning signal across frequencies, emphasizing high-frequency components that are otherwise underrepresented in pixel-space objectives. Building on this, we introduce a simple training strategy that combines frequency and pixel supervision: we first emphasize frequency-domain learning early to capture all frequencies, and then transition to standard pixel-space v-loss for spatial refinement. This balancing mitigates the low-frequency bias of pixel losses and aligns the training signal with the evolving needs of the model. Our approach is conceptually simple, requires no architectural changes, and acts as a drop-in replacement for flow matching losses. Across multiple model scales, it accelerates convergence by up to 40% while consistently improving FID and perceptual fidelity.
Read the full abstract
Computer Vision
InceptionGS: Generative Bootstrapping for Large-Scale Gaussian Splatting under Unstructured View Sampling
Achieving truly immersive large-scale scene digitization necessitates consistent and visually pleasing rendering across all possible viewing perspectives. However, collecting multi-view images covering every fine detail of a large-scale scene is prohibitive due to scene complexity, capture cost, negligence, or accessibility constraints. As a result, the sampled views tend to be highly unstructured – the majority of the scene is well covered yet certain regions inevitably lack sufficient observations. Existing reconstruction based methods are vulnerable to view scarcity while generation based approaches suffer from generalization, controllability, and 3D consistency issues. To address this challenge, we propose InceptionGS, which bootstraps Gaussian splatting by subtly balancing reconstruction and generation. Starting from an initial Gaussian splatting, InceptionGS reasonably rethinks and repairs problematic regions caused by view scarcity while preserving the quality elsewhere, by softly incorporating scene- and view-adaptive generative priors. Extensive experiments on real-world large-scale scenes demonstrate the superiority and broad applicability of our approach in handling unstructured imagery and boosting high-fidelity Gaussian splatting. Please refer to the supplementary video for better visual demonstrations.
Read the full abstract
Chemical Physics
HiPoly: a hierarchical polymer-native AI framework for property prediction and generative design
Polymeric materials are central to modern technologies, with applications ranging from energy to health and transportation. Although AI has made significant advances in materials discovery, the hierarchical structure of polymers across multiple length scales makes them inherently difficult to represent in a unified and physically meaningful way. Here we introduce HiPoly, a polymer-native AI framework that processes complete polymer descriptions through a three-level hierarchical graph architecture built on the G2RINS representation. HiPoly encodes stochastic inter-monomer connectivity, composition, and molecular weight directly within its architecture, using physically motivated design principles that mirror the multi-scale nature of polymeric systems. The framework establishes an end-to-end AI-driven workflow from experimental formulation data to property prediction, generative molecular design, and physics-based validation through molecular simulations, all unified by a single polymer representation. We demonstrate state-of-the-art prediction accuracy for thermophysical properties of multi-component polymer systems, with ablation studies confirming that each hierarchical design choice contributes independently to model performance. As an example, the generative design pathway is applied here to the discovery of sustainable alternatives to persistent fluorinated polymers, where it is possible to identify and independently validate PFAS-free candidates with target surface-energy properties. This work demonstrates how polymer-native AI can accelerate discovery by linking representation, prediction, and design across complex polymer chemistries.
Read the full abstract
Information Retrieval
Incremental Pooled LLM Evaluation for Cost-Effective Retrieval Model Selection
Selecting a retrieval model for a production RAG system requires reliable comparative evaluation, but obtaining relevance judgments at scale is expensive and difficult to repeat as new candidate systems arrive. We study pooled LLM evaluation, in which an LLM judges the union of documents retrieved by the current set of candidate systems, and the pool is then expanded incrementally as new systems are introduced by judging only the new documents they contribute. These judgments are reused to evaluate all systems on a common basis. We validate this approach on four retrieval benchmarks with 11 systems spanning dense, sparse, and hybrid configurations, and deploy it to compare 62 retrieval configurations for a financial news QA system. Pooled LLM rankings correlate strongly with gold-standard evaluation across datasets, and 97% of pairwise system orderings are preserved once bootstrap uncertainty in the qrels is taken into account. In production, document overlap yields 65-80% judgment reuse and up to 4.9x lower evaluation cost, allowing teams to benchmark new retrieval candidates without re-judging previously assessed documents. These results suggest pooled LLM evaluation is a practical and cost-effective workflow for incremental retrieval model selection in deployed systems.
Read the full abstract
Materials Science
Understanding Graphene-Perovskite Interactions: From Flake Chemistry to Crystallisation and Solar Cell Performance
Graphene-derivatives are widely employed materials to improve bulk and interface properties of metal-halide perovskite devices. Yet the implications of their flake chemistry and interactions with the perovskite precursors remain unclear. Here, we show that pristine graphene flakes (GF) and more conventional graphene oxide flakes (GOF) are not interchangeable. Density functional theory calculations reveal that GOF interacts more strongly with the perovskite lattice but induces larger structural distortions, stronger interfacial polarisation, and localised gap states. In contrast, GF forms comparatively non-disruptive contacts, a response retained across a wide compositional range. Machine-learning atomistic simulations further show that GF contacts both Pb- and I-containing regions of solvated perovskite nanocrystals, with a strong solvent dependency. Solution spectroscopic characterization indicates that GF additives serve as scaffold for preorganised Pb/I-containing precursors, favouring film crystallisation. In this sense, GF enhances solar cell performance across perovskite compositions, but particularly those facing a more challenging crystallisation. In mixed Sn-Pb perovskite solar cells, GF raises the champion power-conversion efficiency from 21.5% to 23.7% with improved storage stability. These results establish pristine GF as a chemically defined additive and connect its atomic-scale interactions with precursor organisation, crystallisation, device performance, and stability.
Read the full abstract
Probability
Sharp mean-field estimates for diffusive log/Riesz gases in the Hilbert–Schmidt regime
We study fixed-temperature logarithmic and Riesz gases after subtracting the leading mean-field contribution relative to a prescribed background law. For repulsive interactions in the Hilbert–Schmidt regime, we prove $N$-uniform bounds and quantitative convergence of the resulting modulated partition function to a normalization expressed by the Carleman–Fredholm determinant of the centered interaction operator. We show that the Hilbert–Schmidt threshold is sharp and obtain explicit lower bounds on the rate of divergence at and above it; these rates are expected to be nonoptimal. The proof combines positive-definite truncations and a low/high-frequency decomposition with exponential inequalities and Gaussian-chaos asymptotics for canonical degree-two $U$-statistics. As consequences, we establish entropic commutator estimates with the sharp $O(N^{-1})$ additive scale in the modulated-free-energy method, a static joint linear-statistics central limit theorem, and a dynamical central limit theorem for joint linear statistics at finitely many times. This extends the logarithmic partition-function estimates of the first two authors to the full Riesz Hilbert–Schmidt range and identifies the limiting determinant normalization. For the attractive logarithmic interaction at sufficiently small inverse temperature, we also prove analogous results.
Read the full abstract
High Energy Astrophysical Phenomena
An Exact Engine for Black-Hole Jets
For fifty years the Blandford–Znajek mechanism has been a mechanism and not an exact solution: force-free electrodynamics on a prescribed metric, with the jet's own field carrying no weight. Here that field gravitates. A split monopole weighs as much as a magnetic monopole, so its self-gravity is the magnetic Reissner–Nordström geometry; two copies glued across an equatorial current sheet put hole, disk and jet-driving flux into one spacetime, and force-free plasma makes a magnetosphere of it. Three things then follow that no test-field calculation can see. First, a steady jet is impossible. A stationary horizon cannot be heated but a slipping magnetosphere necessarily heats it, so the only stationary state is a dead one corotating with the hole, and a working jet is a black hole in decay at rates the field equations fix rather than assume. Second, flux enters the laws of black-hole mechanics as a charge, sharing one extremality budget with spin. That budget caps the jet power at $c^5/48G$ whatever the mass. Third, the lifetime output is finite: a maximally spinning hole delivers $1-e^{1/4}/\sqrt2=9.2\%$ of its mass and grows its horizon area by exactly $\sqrt e$. The horizon's own moment of inertia vanishes with the irrational exponent $(\sqrt{17}-1)/2$ set by the near-horizon throat, and what is left is a rotating hole whose angular momentum has passed to its own field.
Read the full abstract
Cryptography and Security
SPADE: SPaT Attack Detection from the Connected Vehicle's Perspective
Signal Phase and Timing (SPaT) messages are a cornerstone of connected vehicle (CV) safety, enabling CVs to perceive and respond to intersection state through Vehicle-to-Infrastructure (V2I) and Vehicle-to-Vehicle (V2V) communication. The integrity of these messages is threatened by a range of application-layer attacks that can bypass conventional authentication when a roadside unit or peer vehicle is compromised. Existing intrusion detection research either defends the infrastructure side or targets V2V Basic Safety Message (BSM) / Cooperative Awareness Message (CAM) misbehavior, leaving the onboard CV perspective on SPaT integrity unaddressed. To close this gap, we introduce SPADE — the SPaT Attack Detection and Evaluation dataset — a labelled, multi-modal, simulation-based dataset designed specifically for deep learning IDS research in this space. SPADE is generated through Eclipse MOSAIC using runtime attack injection at the SAE J2735 application layer across six attack classes and one benign class. By combining four intersection geometries, six operating conditions, and five independent random-seed repetitions, SPADE comprises 180 unique base scenario runs, yielding $\sim$1,890,000 labelled timestep records (270,000 per class). Each record fuses SPaT message fields, onboard camera confidence scores, and cooperative V2V peer data across 40 features, reflecting the multi-modal signal space required to distinguish deliberate attacks from environmental degradation. The dataset, generation code, and scenario configurations are released publicly to support reproducible and comparative IDS research in C-V2X security. The developed toolbox, instructions, and dataset link are publicly available on GitHub: https://github.com/jdinovo/SPADE.