Peer Revenue Play credits. Real papers.

Papers

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

All of arXiv

Wednesday 9 September

Signal Processing

Probabilistic Symbol-Level Precoding based Affine Frequency Division Multiplexing Transmission

Affine frequency division multiplexing (AFDM) has recently gained significant attention due to its robustness against time-frequency doubly selective channel fading. However, the high computational complexity at the receiver poses a critical challenge for practical deployment. To overcome this issue, we propose a probabilistic symbol-level precoding (SLP)-based AFDM transmission framework, in which the processing burden in downlink transmission is shifted from the user to the base station (BS), enabling direct symbol detection without channel estimation or equalization at the receiver. In the proposed framework, the BS exploits the uplink channel state information (CSI) to design the downlink transmit waveform based on uplink-downlink channel reciprocity. In particular, we innovatively introduce a probabilistic SLP technology by explicitly characterizing the likelihood of symbol detection errors under noise perturbations. Specifically, the transmitted symbols are optimized to minimize the likelihood that the received symbols fall into erroneous decision regions, where the resulting error-probability minimization problem is subsequently approximated as a second order cone programming (SOCP) problem by exploiting the monotonicity of the objective function. Simulation results show that the proposed probabilistic SLP-based scheme achieves performance comparable to that of conventional AFDM receivers, whilst enjoying significant reduction of computational complexity at the receiver end. These results demonstrate the effectiveness and practical potential of the proposed approach.
Read the full abstract

All authors Shuntian Tang, Xinyi Wang, Shuangyang Li, Tianqi Mao, Zilong Liu, Zesong Fei

Hardware Architecture

UNISON: A Co-Designed Near-Memory Scheduler of Session KV Residency for LLM Agents

Large language models are increasingly composed into agent loops that plan, call tools, and resume the same task after each action. These loops press a shared memory hierarchy harder than conventional multi-turn chat, because they hold a growing key-value (KV) prefix across tool waits and place many sessions on one SRAM/HBM pool, so that eviction and hierarchical placement become a session-level efficiency problem orthogonal to compute-mode optimization. Existing proxies based on recency, timeout, or identity miss the mechanism information of the loop and therefore treat a live wait as a cold, discardable unit. We present Unified Native Inter-turn Session Orchestration Nexus (UNISON), an event-driven near-memory scheduler in which Survival-Penalty Eviction for Agent Return-gap (SPEAR) and Tiering in Idle-window DMA Events (TIDE) share one live ranking. SPEAR selects who leaves from a gap average and a turn-indexed hazard, while TIDE spends the observed wait as a DMA budget for who sits in the fast tier. On coding and general-mission benchmarks with three model families, totaling 1,415 sessions and 33,596 turns, the joint policy is the best non-oracle entry on every trace, raising hit rate by 0.3% to 23.1%, reducing AMAT by 22% to 51%, and lowering TTFT by 58% to 89% on long-horizon traces. A structural necessity analysis shows that the unified near-memory design cannot be decomposed into independent IPs or realized in software without re-introducing documented failure modes. The 28-nm CMOS scheduling core occupies 0.169 mm^2 at 13.6 mW and 150 MHz, a negligible overhead relative to the KV hierarchy it manages, reproducing the floating-point ranking at Kendall tau exceeding 0.998.
Read the full abstract

Probability

Sharp Typical Distance and Exponential Small-Ball Bounds in One-Step-Cliff Nielsen Geometry

Let $D=2^n$ and equip $\operatorname{PU}(D)$ with the one-step-cliff Nielsen metric, with quadratic metric coefficients one in Pauli directions of weights one and two and $D^2$ in all higher weights. We prove that the distance from the identity of a Haar-random element, normalized by $D$, converges to $\pi/\sqrt3$ in probability and in $L^p$ for every $1\le p<\infty$. Quantitatively, it lies within $O(D^{-1/8}(\log D)^{1/2})$ of this limit outside a set of Haar measure at most $\exp\{-\Omega(D^{7/4}\log D)\}$. For each fixed $0<x<\pi/\sqrt3$, the ball of radius $xD$ has Haar measure $\exp\{-\Theta_x(D^2)\}$; for $x>\pi/\sqrt3$, its complement has measure at most $e^{-c_xD^2}$ for some $c_x>0$. As $x\uparrow\pi/\sqrt3$, the lower and upper logarithmic rates are both asymptotic to $(\pi^2/3-x^2)^2/(16\zeta(3))$. The small-ball upper bound follows from a comparison of Jacobi determinants, obtained by rescaling the linearized geodesic equations and applying Kato transport. Weyl integration reduces the remaining integral to an Abel-regularized logarithmic-energy estimate on the circle. A centered principal logarithm and concentration of the circular unitary ensemble eigenangle second moment give the distance upper bound.
Read the full abstract

Programming Languages

UnsafeChecker: Finding Soundness Bugs in Rust Safe Abstractions

Rust guarantees memory safety without garbage collection through a strict ownership and borrowing system. However, for low-level systems programming, many widely used libraries rely on the unsafe keyword. These libraries encapsulate raw-pointer operations behind safe APIs to form safe abstractions. A single mistake in this internal unsafe code can break its safety contract, rendering the abstraction unsound and allowing safe clients to trigger undefined behavior. Detecting these potential soundness violations is challenging. Existing static analysis tools for C/C++ ignore Rust-specific safety contracts, while current Rust tools lack the deep semantic modeling required to track the contexts that raw pointers erase. To address this gap, we present UnsafeChecker, a compiler-integrated static analysis framework for detecting potential soundness violations in Rust safe abstractions. UnsafeChecker analyzes Rust MIR using a flow-sensitive abstract interpretation that maintains a shared state with three components: ownership, object validity, and layout. Each warning rule consumes the subset of facts needed for the corresponding Rust safety obligation. UnsafeChecker reports both instruction-level undefined behavior and boundary-level contract violations that may escape through safe APIs. We evaluate UnsafeChecker on a benchmark of 46 RustSec vulnerabilities, which contain 53 ground-truth bugs. UnsafeChecker outperforms several state-of-the-art tools, detecting 32 CVEs and covering 36 bugs (67.9% recall) with 51.6% alert-level precision. Furthermore, in a large-scale scan of real-world crates on crates.io, UnsafeChecker uncovered 114 previously unknown bugs across 83 crates, with 45 confirmed and 27 already fixed by maintainers.
Read the full abstract

All authors Xizhe Yin, Yaokun Zhang, Yang Feng, Baowen Xu

Instrumentation and Detectors

Spectral Discrimination of Deposited Gamma-Ray Energies in a Simulated CeBr$_3$ Scintillator

We show that wavelength measurements of individual detected optical photons may provide additional information about gamma-ray energy deposited in a CeBr$_3$ crystal when the detected-photon-count distributions overlap for nearby gamma-ray energies. Monoenergetic 662 and 629 keV gammas are used in a Geant4 simulation of a $25\times25\times20~\mathrm{mm^3}$ CeBr$_3$ crystal. Assuming a light yield of $6.0\times10^4$ photons/MeV, a wavelength-independent photon-detection efficiency of 30%, and a wavelength resolution of $\sigma_{\lambda}=40$ nm, we find that the fraction of photons reconstructed above 385 nm gives an event-level separation of $\sim$ 2 standard deviations between the 662 and 629 keV event populations selected within the same $\sim$ 1%-wide detected-photon-count interval. No timing or reconstructed interaction-position information is used. The result demonstrates, within the present simulation model, that event-dependent optical spectra can retain energy information beyond an undifferentiated photon count.
Read the full abstract

Metric Geometry

The topology of Gromov–Hausdorff space

We prove that the Gromov–Hausdorff space is homeomorphic to the real Hilbert space $\ell^2$. We construct a continuous assignment of full-support probability measures that is equivariant under isometries and finite-dimensional local approximations that control all pairwise distances. These approximations yield the absolute retract property for all metrizable spaces. We also prove that any countable family of continuous maps from compact metrizable spaces can be approximated, with respect to a prescribed open cover, by maps whose images form a discrete family.
Read the full abstract

Metric Geometry

The mean distance to a simple closed curve on the sphere

Kimberling's Problem 10 asks for a simple closed curve of prescribed length $L$ (in particular, $L=4\pi$) on the unit sphere minimizing the mean geodesic distance $\mathcal{J}$ from a point of the sphere to the curve. For a positive integer $n$, put $\vartheta_{n}=\pi/(2n)$ and $L_{n}=2\pi/\sin\vartheta_{n}$. We show that the minimum of $\mathcal{J}$ over rectifiable simple closed curves of length at most $L_{n}$ equals $\vartheta_{n}-\tan(\vartheta_{n}/2)$, that it is attained only by curves of length exactly $L_{n}$, and that the sphere-filling ropes $\beta^{n,k}$ of Gerlach and von der Mosel attain it. Kimberling's case is $n=3$: at $L=4\pi$ the minimum is $\pi/6+\sqrt{3}-2=0.255649\ldots$, attained by an explicit six-arc curve and by its mirror image. For $L\le2\pi$ we determine $J(L)$, the infimum of $\mathcal{J}$ over curves of length $L$, exactly: it equals $\pi/2-L/(2\pi)$, attained precisely by the circles of length $L$. At the lengths $L_{n}$ we do not classify all minimizers, but show that every one of them bisects the sphere into two disks of area $2\pi$ and inradius $\vartheta_{n}$ whose inward collars have the largest possible area at every depth. The great circle is the only minimizer for $n=1$, and the $\beta^{n,k}$ are, up to congruence, the only ones of thickness at least $\sin\vartheta_{n}$. For arbitrary $L$ the function $J$ is nonincreasing, and together with the above this brackets it between two explicit values.
Read the full abstract

Mathematical Physics

Counterexamples to GNS Conjecture 1.8

We disprove both parts of Conjecture 1.8 of Gabrielov, Novikov, and Shapiro (GNS). For the Coulomb potential, a generic rational configuration of 24 unit charges in $\mathbb{R}^3$ has at least 18 nondegenerate critical points of Morse index one but exactly 14 effective Voronoi one-cells. A separate proper-line construction has at least two nondegenerate minima but only one relatively effective zero-dimensional cell intersection. For each positive subspace dimension, product suspension gives a proper-subspace counterexample.
Read the full abstract

Analysis of PDEs

Analysis and stability of the FitzHugh-Nagumo system with rapidly oscillating Neumann boundary condition

In this paper we study the FitzHugh-Nagumo system posed on the half-line with a sinusoidal forcing term acting at the origin through the Neumann boun\-da\-ry condition. By applying an averaging method, we show that, if the frequency of the oscillating forcing is high enough, then the system can be approximated by a combination of a highly-oscillatory term and the solution of a simpler system. This approach allows us to establish stability results under persistent time-varying boundary conditions. Along with the averaging method, we use an adaptation of the contracting rectangles method, tailored to handle parabolic equations with spatially and temporally varying coefficients.
Read the full abstract

Fluid Dynamics

Macroscale vortex impingement at a porous-fluid interface induces local heat-transfer enhancement

Externally generated wakes can impinge on porous layers in several practical heat transfer applications. However, the transport of these macroscale vortices and their resulting influence on heat transfer in the porous layer remain unknown. In this paper, two-dimensional pore-resolved simulations are used to examine a square-bluff-body wake impinging on a porous layer composed of an in-line array of heated square obstacles at Re = 500-4000, Pr = 0.7 and 7.0, and porosities $\phi$ = 0.75, 0.85, and 0.95. The incident macroscale vortices break down starting immediately at the porous-fluid interface. Following macroscale turbulent thermal transport in this entrance region, fluctuations farther downstream are aligned with the pore geometry and are generated at the microscale level by pore-throat shear, separation, and repeated obstacle wakes. Direct impact locally improves heat transfer at the porous-fluid interface by increasing the local Nusselt number relative to a non-impingement region. The interfacial heat-transfer enhancement varies with porosity, Reynolds number, and Prandtl number, with a peak enhancement of 18.2% observed for $\phi$ = 0.85, Pr = 7.0, and Re = 1000. The wake-associated macroscale turbulent heat-flux contribution decays within a consistence entrance region extending approximately 3-4 unit cells into the porous layer across all investigated porosities, Reynolds numbers, and Prandtl numbers. For Re$\ge$1000, the persistent wake momentum deficit subsequently causes the heat-transfer contrast to become negative, resulting in lower heat transfer in the impingement region than in the non-impingement region.
Read the full abstract

All authors Thibaut K. Kemayo, Vishal Srikanth, Justin Courter, Rodrigo R. Caballero, Andrey V. Kuznetsov

Computer Vision

LightMedSeg-ISLES: Stroke Lesion Segmentation with 81x Fewer Parameters than nnU-Net

Large networks and ensembles often lead medical image segmentation challenges, but their storage and inference demands complicate deployment. We present LightMedSeg-ISLES, a 1.26-million-parameter pipeline for T1-weighted stroke lesion segmentation in ISLES'26. On a 146-case held-out cohort, flip test-time augmentation produces 0.618 mean Dice and 0.599 lesion-wise F1. A 102.35-million-parameter nnU-Net ResEnc-L produces 0.634 Dice and 0.544 lesion-wise F1 after size filtering. LightMedSeg therefore retains 97.5% of nnU-Net's Dice with 81.4$\times$ fewer parameters while improving lesion-wise F1 by 0.055. Its four-pass TTA operating point requires 4.7$\times$ fewer FLOPs per standardized patch than nnU-Net. It also slightly exceeds filtered UNETR++ and nnFormer. Longer training and stronger augmentation add 0.0358 Dice without increasing capacity, establishing a strong single-checkpoint alternative to much larger models.
Read the full abstract

All authors Giorgi Nikvashvili, Hanxue Gu, Jie Bao, Kang Wang, Yang Yang

Algebraic Geometry

Representation Varieties of Stacks and Trace Maps

We define a derived stack $\mathscr{Rep}_n(X)$ which generalizes the assignment $A\leadsto \operatorname{Rep}_n(A)$ to an algebra of its derived $GL_n$-representation variety of arXiv:1112.1449 from algebras $A$ to perfect stacks $X$ over characteristic 0 fields. In the case of a quasi-projective classical scheme $X$, we show that $\mathscr{Rep}_n(X)$ admits a subfunctor $\operatorname{Quot}^{n,\text{fr}}_{\mathscr{O}_X^n}(X)\subset \mathscr{Rep}_n(X)$, which is in fact represented by a derived scheme almost of finite type. We further construct a Fourier-Mukai integral transform between the derived categories of quasi-coherent sheaves on $X$ and $\operatorname{Quot}^{n,\text{fr}}_{\mathscr{O}_X^n}(X)$ which induces a trace map at the level of Hochschild homology generalizing the trace morphism constructed in arXiv:1112.1449 (for finitely presented commutative algebras). We show that the subfunctor $\operatorname{Quot}^{n,\text{fr}}_{\mathscr{O}_X^n}(X)$ is a derived enhancement of the framed locus of the Quot scheme of points and that this derived scheme is a $GL_n$-torsor over the stack of coherent length $n$ torsion sheaves. Hence, this stack is an analog of the derived character stack for quasi-projective schemes, and we show that for smooth, Calabi-Yau $X$, it inherits a shifted symplectic structure in the sense of arXiv:1111.3209 from the one constructed in arXiv:1812.11913 on the moduli stack of perfect complexes with proper support. This structure is shown to give a generalization of the classical symplectic structure on character varieties of surfaces of genus 1 constructed by Goldman \cite{gold}.
Read the full abstract

Commutative Algebra

Stabilization index of V-number of powers of edge ideals of graphs

We provide a combinatorial description of the stabilization index of the $v$-function of powers of edge ideals for arbitrary graphs.

Instrumentation and Methods for Astrophysics

A pilot study on the CSST astrometric capability: Detecting astrometric binaries with Gaia synergy via simulated data

Context. The China Space-station Survey Telescope (CSST) will provide deep, wide-field epoch astrometry during its 10-year mission. Astrometric binary orbits constrain the masses of stellar and compact-object components. Orbital recovery depends on astrometric precision and temporal coverage. Combining CSST and Gaia data extends the baseline and improves binary detection. Aims. We evaluate CSST, Gaia, and joint astrometry for binary-candidate selection and 12-parameter (12p) orbit fitting at faint magnitudes ($g>17.8$). We also test how regular CSST cadences affect the yield of 12p fits satisfying our criteria. Methods. We constructed a mock catalog, simulated CSST and Gaia epoch astrometry, and fitted five-parameter (5p) single-star models to derive astrometric diagnostics, proper-motion anomaly features, and observational-sampling features. A four-stage histogram-based gradient-boosting classifier used these features to select candidates for 12p orbit fitting and assessment. Results. On the independent test set, the classifier reaches a precision of 0.802 and a recall of 0.181 among eligible true binaries. In the scenario-specific fitted samples, joint astrometry raises the fiducial fraction from 6.76% for Gaia alone to 10.46%; for fitted binaries with $P_{\rm true}>15{\rm yr}$, it rises from 2.37% to 6.78%. The current CSST schedule yields few fiducial fits, while idealized regular cadences increase the yield mainly at $g\lesssim21$. Conclusions. In the simulation, joint CSST and Gaia epoch astrometry yields higher fractions of fitted unresolved binaries satisfying the stated criteria than Gaia-only solution. A practical strategy is to select candidates from 5p diagnostics and astrometric anomalies, obtain more regular CSST follow-up observations, and then fit 12p orbital models and apply the selection criteria.
Read the full abstract

All authors Shangyu Wen, Shilong Liao, Zhaoxiang Qi, Zhensen Fu, Xiyan Peng, Ye Ding, Qiqi Wu, Qi Xu, Xun Sun, Keyu Zhu, Yong Yu

Robotics

JEPA Policy: Diffusion-Free Imitation Learning via Paired Action and Future Representation Prediction

Standard behavior cloning supervises actions without explicitly constraining the future representation paired with each demonstrated action chunk. We introduce JEPA Policy, a diffusion-free framework that uses the action chunk and its observed future representation as paired training targets. Action and future-representation tokens interact in a shared Transformer and are refined through two forward passes. Future prediction can therefore shape the representation used to generate actions. Dual-branch and gradient-routing controls attribute the gain to this shared topology rather than to an auxiliary prediction head alone. Across nine simulated tasks, JEPA Policy improves mean success over the action-only MIP baseline and outperforms Diffusion Policy under the evaluated configurations, while adding 0.29 ms to MIP's model latency. A five-task, 630-episode physical-robot study produces the same pooled ranking. Further audits find no complete representation collapse under action supervision and identify a task-conditioned failure-ranking signal in future-prediction error. These results support paired future-representation supervision as a practical approach to low-latency visuomotor imitation without iterative generative sampling.
Read the full abstract

All authors Jie Xu, Kangjin Yu, Ziyi Jin, Junjie Gao, Liqing Chen, Yixian Li, Shuai Tian, Zhongpu Xia

Quantum Physics

A Sum-of-Squares Hierarchy with Quadratic Convergence for Quantum Channel Coding

Computing the optimal success probability for transmitting classical messages through a single use of a quantum channel is NP-hard, even for two messages. An existing semidefinite programming hierarchy based on symmetric extensions provides convergent upper bounds with an a priori error estimate that decays as the inverse square root of the extension level. In this work, we construct a Hermitian sum-of-squares hierarchy for an arbitrary number of messages and prove quadratic convergence in its level. The error bound is proportional to the advantage over random guessing. Our approach combines state-discrimination duality with positive polynomial kernels on products of spheres to construct feasible polynomial dual certificates. For binary messages, the resulting bounds give a multiplicative approximation from above of the trace-norm contraction coefficient.
Read the full abstract

Multiagent Systems

Who Are They to Each Other? Multi-Agent Reasoning for Speaker Relationship Inference

Inferring speaker relationships from spoken conversations is an important step towards socially aware speech understanding. However, this task remains underexplored, and supervised modeling is costly to train and scale. At the same time, existing inference-time LLM approaches provide limited structure for handling subtle, distributed, and multimodal relational cues that may support multiple plausible interpretations. To address these limitations, we introduce a training-free multi-agent reasoning framework that organizes inference through structured interaction among LLM agents, allowing relationship judgments to be proposed, challenged, and adjudicated without task-specific training. We instantiate this framework with two complementary designs. We propose Multi-Role Multi-Agent Debate as a task-specific adaptation of standard multi-agent debate for speaker relationship inference, assigning agents complementary roles or social-theory-grounded perspectives rather than a single undifferentiated viewpoint. In contrast, we introduce Multi-Agent Compete, a competition-based protocol that compares agent judgments through pairwise adjudication, eliminates weaker candidates, and retains the most defensible one. We evaluate these methods on the Seamless Interaction dataset across different modality settings, covering both binary classification and fine-grained relationship-detail prediction. Results suggest that they improve over zero-shot and existing multi-agent baselines in most cases. Human evaluation further suggests that this task is challenging even for people. LLM methods can sometimes outperform human annotators in text-included settings but are less competitive in the audio setting. Together, these findings suggest that relationship inference benefits from structured inference-time interaction among agents, while acoustic cues are not yet fully captured by current models.
Read the full abstract

All authors Yaohan Guan, Yen-Ju Lu, Yuzhe Wang, Junhyeok Lee, Jesus Villalba, Laureano Moro Velazquez, Thomas Thebaud, Najim Dehak

Artificial Intelligence

Seven Sources of Physical AI Capability Formation

Capabilities relevant to Physical AI can arise from materially different formation histories, yet existing taxonomies organized by morphology, architecture, learning algorithm, task, or domain do not directly answer what gives rise to a capability. We define a capability-formation source as a factor materially contributing to capability formation, distinct from components or construction steps. We identify seven non-exclusive sources: Recorded-Experience (RE), Predictive-Modeling (PM), Evaluative-Interaction (EI), Surrogate-Environment (SE), Mechanism-Grounded (MG), Embodied-Coupling (EC), and Evolution-Driven (ED) Formation. Using reconstructive induction with theoretical saturation, we traced a research matrix to primary studies, deduplicated the literature, set coding rules, and conducted three rounds of maximum-difference and negative-case sampling. Challenges included curriculum and self-supervised learning, active inference, open-ended and developmental learning, planning and search, neuro-symbolic architectures, digital twins, generative physical world models, and morphology-control co-design. Within the scope and criteria fixed as of September 4, 2026, all 49 evidence records were explainable by the seven sources individually or in combination. No R1-R3 challenge produced an irreducible eighth source, and R3 required no new core definition or substantive boundary rule. We therefore claim theoretical saturation within the stated scope, not logical completeness or exhaustive future coverage. The framework distinguishes similarity in observed capability from similarity in how it was formed, supporting analysis of explanation, transfer, replication, dependencies, governance evidence, and geoeconomic foundations.
Read the full abstract

Computer Vision

Hyperbolic Geometry for Open-World Object Detection in Remote Sensing Imagery

Open-world object detection (OWOD) extends closed-set detection by requiring models to identify unknown objects and incrementally learn them once annotations become available. In remote sensing imagery, object categories often exhibit latent hierarchical relationships that may be inadequately represented in the Euclidean spaces commonly adopted by existing methods, limiting unknown-object recall and incremental-learning performance. To address this issue, we investigate hyperbolic geometry for OWOD in remote sensing imagery and propose HyRS-OWOD. To improve unknown object recall, we design a two-step unknown-object discovery mechanism: a Decoupled Objectness Learning (DOL) module that disentangles foreground perception from semantic information to separate foreground proposals from background regions, followed by a Hyperbolic Uncertainty Learning (HUL) component that leverages the radius of hyperbolic embeddings as an uncertainty-aware cue for known-unknown discrimination. For incremental learning, we develop a Hyperbolic Metric Learning (HML) strategy that enhances inter-class separability, facilitating the incorporation of novel categories while mitigating catastrophic forgetting. Experiments on three remote sensing benchmarks demonstrate consistent improvements in unknown recall and incremental learning over state-of-the-art OWOD methods.
Read the full abstract

All authors Wuzhou Li, Jiawei Zhou, Shenghang Wang, Xiang Li

Artificial Intelligence

From State Synchronization to Cognitive Self-Evolution: An Operational Architecture for Cognitive Digital Twins

As Digital Twin (DT) systems evolve beyond state synchronization toward task-oriented and knowledge-driven operation, Cognitive Digital Twins (CDTs) have emerged as an extension that incorporates cognitive capabilities into twin operation. Existing CDT studies often focus on specific enabling techniques, such as learning modules, knowledge graphs, and large language models, while providing limited insight into how cognition can be systematically integrated into DT architectures. To address this issue, this paper proposes a four-layer CDT architecture consisting of the physical layer, digital-twin layer, cognitive layer, and task layer. The proposed architecture establishes a self-evolving closed operational loop spanning these four layers, in which physical states are synchronized into digital representations, cognition constructs task-specific cognitive models through knowledge, memory, and attention, and task-level decisions are generated under practical constraints. Operational feedback further refines cognitive experience and updates relationships and annotations in the digital representation, enabling subsequent task interpretation, initiation, and reasoning to evolve with system operation. Based on this framework, two representative operation modes are characterized: user-request-driven cognition and self-driven cognition. We further discuss key enabling mechanisms and deployment challenges associated with semantic communication, knowledge querying, task orchestration, and closed-loop synchronization. A lightweight simulation study illustrates reliable closed-loop task feasibility under limited semantic information and improved operational efficiency through accumulated task experience. The proposed framework provides a structured foundation for the design and development of future CDT systems.
Read the full abstract

All authors Haoran Gao, An Li, Zhen Li, Jun Cai

Networking

Contextual Bandit-Based Decomposition of Network Slice Requirements under Cumulative Resource Budget Constraints

End-to-end (E2E) network slices (NSs) are provisioned across multiple domains of the 5G network. In hierarchical NS management, a tenant submits a network slice request (NSR), which specifies E2E service level agreement (SLA) requirements. Rather than managing these domains directly, an E2E controller decomposes each NSR into domain-level SLA requirements and delegates resource allocation to domain-specific controllers, which return feasibility and resource-consumption feedback. A poor decomposition policy can therefore cause rejection of the current request by producing infeasible requirements or reduce future admission opportunities by concentrating resource consumption in bottleneck domains. We call this decomposition-policy optimization problem the network slice request decomposition problem (NSR-DP). For practical operation, online approaches to NSR-DP have been proposed. Such approaches must jointly meet two requirements: (R1) control long-term resource budgets and (R2) adapt each decomposition to the performance targets and guarantee levels specified in the arriving NSR's SLA. To meet these requirements, we introduce contextual constrained kernel bandits (CCKB) as an online solution for NSR-DP. To address (R1), CCKB raises penalties for using resources that become tight, thereby discouraging decompositions that consume bottleneck resources. To address (R2), it uses Gaussian processes (GPs) to predict, for the current NSR, the reward and resource usage of candidate decompositions, allowing it to select a decomposition suited to the performance targets and guarantee levels. We establish high-probability guarantees for the resulting formulation and show through extensive 5G simulations across topology, bottleneck, and traffic-mixture settings that CCKB outperforms the baselines in the large majority of conditions.
Read the full abstract

All authors Masaki Kobayashi, Akito Suzuki, Ryoichi Kawahara, Masahiro Kobayashi

Cryptography and Security

Scalable Composition of Byzantine Agreements under Reorder Attacks

Byzantine agreement (BA) is a foundational building block in distributed systems, and the security analysis of BA protocols under multi-instance executions has attracted increasing attention. However, most existing adversary models focus solely on party corruption and neglect important threats posed by adversarial manipulations of communication channels in the network. Through channel attacks, messages can be reordered across multiple executions and lead to violations of the protocol's security guarantees, In this work, we present the first adversary model that combines party corruption and channel attacks. Based on this model, we establish new security thresholds for Byzantine agreement under parallel and concurrent compositions, supported by complementary impossibility and possibility results that match each other to form a tight bound. For the impossibility result, we show that even authenticated Byzantine agreement protocols cannot be secure under parallel composition when $n \leq 3t$ or $n \leq 2c + 2t + 1$, where $t$ and $c$ denote the number of corrupted parties and communication channels, respectively, and $n$ is the number of parties. For the possibility result, we prove the existence of secure protocols for unauthenticated Byzantine agreement under parallel and concurrent composition, when $n > \max\{3t, 2c+2t+1\}$. We first provide general black-box compilers that transform any single-instance secure BA protocol into one that is secure under parallel and concurrent executions without additional security assumptions. To optimize performance, we further design refined compilers using erasure-correcting codes. These refined versions significantly reduce communication overhead, particularly for long messages, where they achieve a constant multiplicative overhead compared with the original protocol, thus achieving the same asymptotic communication complexity.
Read the full abstract

All authors Jing Chen, Jin Dong, Jichen Li, Xuanzhi Xia, Wentao Zhou

Data Structures and Algorithms

Streaming Algorithms for Gaussian Kernel Density Statistics

Motivated by data produced by generative systems, \cite{LZ26b} formulates similarity-aware statistics via a weighted similarity graph, replacing equality with similarity in classical frequency-based statistics. Although this framework captures semantic relationships between nonidentical items, under general similarity functions even coarse one-pass approximation can require linear space. We therefore ask whether the geometric structure present in natural vector similarities can overcome this barrier. We answer this question affirmatively for the Gaussian kernel. For fixed-dimensional Euclidean vector streams, we study similarity-aware analogues of classical frequency statistics, including the number of distinct elements and frequency moments, through the diversity index and Gaussian density moments. We give one-pass sublinear-space approximation algorithms that exploit the geometric and analytic properties of the Gaussian kernel, and complement them with lower bounds. Our results show that geometric structure can fundamentally change the streaming complexity of similarity-aware statistical analysis.
Read the full abstract

Combinatorics

Extending Symmetric Layer-Rainbow Latin Cubes

An $n\times n\times n$ array on $n^2$ symbols is a layer-rainbow Latin cube if every layer contains every symbol exactly once. We call it symmetric if $L_{ij\ell}=L_{j\ell i}=L_{\ell ij}$ for distinct $i,j,\ell$ and $L_{iij}=L_{jji}$, $L_{iji}=L_{jij}$, $L_{ijj}=L_{jii}$ for distinct $i,j$. We determine exactly when a symmetric layer-rainbow Latin cube of order $m$ embeds in one of order $n$, giving a three-dimensional analogue of Cruse's embedding theorem. Call a positive integer admissible if it is congruent to $0$ or $2$ modulo $3$, with $1$ admissible and $3$ excluded. For $n>m$, an embedding exists if and only if $m,n$ are admissible, $(m,n)\ne(2,5)$, and \[ \begin{cases} n\geq2m,&n-m\not\equiv1\pmod3,\\[1mm] \displaystyle n\geq m+\frac{\sqrt{48m^2+1}-1}{6},&n-m\equiv1\pmod3. \end{cases} \] Via the equivalent one-factorization problem for a non-uniform hypergraph, fair detachment reduces the proof to an exact integer allocation. We also determine the structure forced at both sharp boundaries and obtain infinitely many equality cases. At order eight, we construct a symmetric layer-rainbow Latin cube admitting the natural diagonal action of $\operatorname{PSL}(2,7)$, whose induced action on the $64$ symbols has orbit sizes $1,7,28,28$.
Read the full abstract

Statistical Mechanics

Renewal process's guide to fractional Navier-Stokes equations

The Navier-Stokes equations, which remain unsolved, are crucial equations in fluid mechanics. Discovering the solutions to the Navier-Stokes equations is one of the challenging Millennium problems. In 1900, Hilbert proposed a potential approach to tackle this problem by establishing the relationship between microscopic dynamics and the macroscopic continuum equations. The key bridge is the derivation of Boltzmann equation and the theory of probability. In this paper, we shall use the collision renewal process with arbitrarily distributed waiting times to derive the Boltzmann equation for the time evolution of the probability of the velocity and the displacement of the particle, based on which we prove that the renewal process with exponential collision waiting time is equivalent to the classical Navier-Stokes equations, and that with power-law waiting time is equivalent to the fractional Navier-Stokes equations. Since the collision renewal process with arbitrarily distributed waiting times is a random process and is easy to perform the stochastic simulations of trajectories to obtain the corresponding solution, we actually find a stochastic approach to solve the classical and fractional Navier-Stokes equations.
Read the full abstract

Functional Analysis

Sharp Conditioning for Matrix Recovery by Finite Affine Orbits

Let \(q=p^h\) be an odd prime power, and let the affine group \(\mathbb F_q\rtimes\mathbb F_q^\times\) act through its canonical \((q-1)\)-dimensional irreducible representation. Qualitative matrix recovery for these rank-one orbits is known. We determine sharp lower singular-value bounds for explicit real generating windows. First, we compute the exact least singular value for every nonnegative two-level window over \(\mathbb F_{p^h}\) with \(h\geq2\), and identify the unique optimizer when \(q-1\geq10\). The resulting conditioning stays bounded away from zero on every fixed odd-characteristic tower and is within an explicit characteristic-dependent factor of the best possible value over all real windows. For every \(q=3^h\), \(h\geq2\), we construct a real three-level absolute-trace window (constant on the fibers of \(\operatorname{Tr}_{\mathbb F_q/\mathbb F_3}\)) whose least singular value is \[ \frac{q(\sqrt2-1)}{q(2-\sqrt2)-1}>\frac1{\sqrt2}. \] For the full class of real windows constant on the three trace classes, we reduce the least singular value to four scalar expressions and a symmetric \(2\times2\) matrix. This yields the exact global optimum and all equality cases: the displayed trace window is uniquely optimal up to global sign and interchange of the two nonzero trace classes. Thus zero trace mean follows from optimality. The proof combines an explicit sum-of-squares identity, uniform quadratic-form certificates, and a finite-geometric block decomposition of the orbit measurement operator.
Read the full abstract

Solar and Stellar Astrophysics

Planetary Nebula Central Stars as Tracers of Planetary Nebula-Star Cluster Associations in the Galaxy

Planetary nebulae (PNe) that are physically associated with star clusters are rare but crucial for stellar evolution studies because they allow their initial and final stellar properties to be determined independently, contributing to constraints on the initial-to-final-mass relation (IFMR), which traces stellar mass loss. Although the identification of such pairs is essential, most reported cases have been identified incidentally rather than through systematic searches. Here, we attempt to uncover new such associations in our Galaxy employing literature PN central stars (CSs) and star cluster catalogues. Exploring relevant angular positions on the sky and kinematic Gaia DR3 data, we identify 38 candidate CS-cluster pairs. The components of eight of these pairs present consistent interstellar extinctions, and their CS properties are compatible with stellar evolution models. Two of these candidate pairs are already reported in the literature as true associations. We evaluate separately the remaining six pairs using literature PN observational data, assessing the likelihood of each pair being a true association. The initial and final masses of our best candidates are mostly consistent with the latest IFMR predictions and confirmed cluster-PNe points. An additional 46 CS-cluster pairs have components with compatible angular positions but lack kinematic CS data and are thus presented separately. More data are needed for a precise estimation of the physical parameters of most of our candidates and to confirm these associations, which will be addressed in future studies.
Read the full abstract

All authors Vasiliki Fragkou, Luis Lomelí-Núñez, Dante Minniti, Arianna Cortesi, Denise R. Gonçalves

Mathematical Physics

Hadamard–Tetrahedral Symmetric Spacetime and Clifford-Like Factorizations of Dirac and Maxwell Operators

A symmetric representation of Minkowski spacetime is constructed by combining a normalized fourth-order Hadamard transformation with an imaginary temporal coordinate. The resulting complex coordinates $q^a$ distribute time and space equally among four components and transform the Minkowski metric into a Euclidean-form complex bilinear metric. The four Hadamard spatial sign vectors form a regular tetrahedron and appear simultaneously in the transformed Dirac and Maxwell operators. The Dirac matrices satisfy a Euclidean-form Clifford algebra and contain tetrahedral combinations of Pauli matrices, whereas the Riemann–Silberstein Maxwell matrices realize the corresponding spin-1 structure. A gauge-invariant QED Lagrangian is formulated in the same coordinates. In Fourier space, the Dirac and Maxwell systems are governed by the common invariant $\kappa_a\kappa^a$, while the complete Maxwell operator admits an exact rectangular Clifford-like factorization. The formulation is further extended to the weak-field Einstein–Dirac–Maxwell system. In harmonic gauge, the linearized gravitational field is governed by the same scalar wave operator $\Box_q$, while gravitational corrections to the Dirac and Maxwell equations enter through the perturbed tetrad, spin connection, and curvature. This common principal-operator structure suggests shared spectral kernels and scalar-wave preconditioning for coupled field solvers.
Read the full abstract

Operator Algebras

Pullbacks of Groupoid $C^*$-Algebras over Disjoint Invariant Sets

We establish a pullback theorem for \(C^*\)-algebras of locally compact Hausdorff étale groupoids. The theorem shows that, when the unit space is covered by two closed invariant subsets whose complements are disjoint open invariant subsets, the corresponding groupoid \(C^*\)-algebras form a pullback diagram in the category of \(\mathbb T\)-\(C^*\)-algebras and \(\mathbb T\)-equivariant \(*\)-homomorphisms, for the gauge actions induced by a \(\mathbb Z\)-valued cocycle and its restrictions. We then prove a collection of boundary-path decomposition theorems for graphs, relative graphs, and topological graphs. We show that admissible decompositions of graphs, together with their analogues in the relative and topological settings, induce corresponding decompositions of boundary path spaces. Combining these decomposition theorems with the groupoid pullback theorem, we recover previously known pullback theorems for graph \(C^*\)-algebras, relative graph \(C^*\)-algebras, and topological graph \(C^*\)-algebras. Thus these pullback phenomena are explained by a single groupoid-theoretic mechanism.
Read the full abstract

Superconductivity

Eight-unit-cell electronic modulations in cuprates originating from local molecular orbitals

The pair density wave (PDW) state with eight-unit-cell (8a0) periodicity has been widely regarded as the primary order in cuprates, yet its existence and origin remain subjects of intense debate. Using spectroscopic imaging scanning tunneling microscopy, we observe spatial modulations of the electronic states with approximately 8a0 periodicity in both the superconducting and insulating regimes of hole-doped Ca2CuO2Cl2 cuprate. We find that the 8a0 spatial patterns are generated by the formation of molecular orbitals by doped holes, which organize into 4a0*4a0 plaquettes as the basic unit. Our results identify the 4a0 molecular orbital as the fundamental electronic building block in cuprates, while the 8a0 PDW represents a spatial subharmonic that emerges at sufficiently high doping.
Read the full abstract

All authors Zhiheng Yao, Sixuan Chen, Jianfa Zhao, Shusen Ye, Weixiang Qu, Ning Xia, Yuling Dai, Luchuan Shi, Hongrui Zhang, Zhenqi Hao, Changqing Jin, Shuo Yang, Yayu Wang