arXiv subjects

Peng Xu

Publications and source records attributed to Peng Xu.

At least 19 recordsLinked to original sources

arXiv2026

USDB: Scaling Storage for Next-Generation Recommender Systems at ByteDance

Recommendation models consume increasingly long, attribute-rich user histories, making small continuous updates and repeated large reads a central storage asymmetry. We present USDB, ByteDance's unified sequence storage practice for streaming and batch workloads. CascadeLSM time-based compaction stages materialize application-specific histories; differentiated replicas share persistent files; partial dumps reconstruct inference inputs from captured recent state and versioned history; and a hybrid columnar format accelerates projected reads. These designs reuse maintenance results across ingestion, inference, and training. Production migration records show 85.8% lower mean write latency, while normalized deployment accounting shows 42.1% lower infrastructure cost. Controlled comparisons with an Optimized RocksDB sequence baseline show faster long-history reads across projections and client loads. Aggregation pushdown sustains 8.5x the throughput of ordinary reads under a constrained receive budget. Production reconstruction and direct-file training results demonstrate how shared sequence maintenance supports fresh serving and repeated historical access.

cs.DB↗

arXiv2026

Probing Hidden Dimensions with Multimessenger Gravitational Waves

The dimensionality of spacetime is a fundamental property of nature, yet whether our observed four-dimensional spacetime is embedded in a higher-dimensional geometry remains experimentally unresolved. Gravitational waves (GWs) offer a direct probe of compact extra dimensions through the strong-field dynamics of binary coalescences. In this work, we investigate extra-dimensional modifications to GW generation within a compactified Kaluza--Klein framework. We derive the Newtonian conservative binary dynamics and the leading-order extra-dimensional correction to the GW energy flux and formulate the corresponding waveform modifications within the parameterized post-Einsteinian (ppE) framework. Combining these results with the previously derived 1PN amplitude correction arising from perturbations of the internal volume, we construct modified GW waveform templates for ground- and space-based detectors. To alleviate parameter degeneracies in GW parameter estimation, we combine the observed strain data of GW170817 with independent electromagnetic observations in a multimessenger Bayesian analysis, yielding a 95\% credible upper bound of $n<0.11$ on the phenomenological dimensional parameter. We further investigate injected multimessenger massive black hole binary observations with Taiji, finding that the upper bound could be improved by nearly two orders of magnitude to $n<0.00178$. These results highlight the potential of multimessenger GW observations to probe extra-dimensional effects and test fundamental gravity in the strong-field and dynamical regime.

gr-qc↗

arXiv2026

On the mapping between bound states and black hole quasinormal modes via analytic continuation: a spectral instability perspective

In this work, we investigate the relation between bound states and quasinormal modes within black hole perturbation theory in the context of spectral instability. Our analysis indicates that the reliability of such spectral mapping stretches beyond the domain of validity of the analytic continuation employed to connect the perturbative bound-state problem to the corresponding open-system dynamics. However, for the numerical scheme proposed by Völkel to work, the transformations of the metric parameters must be carried out in a region where the underlying Taylor expansion is convergent. As analytically accessible explicit examples, we explore the perturbed delta-function and Pöschl-Teller potential barriers. For the latter, we construct two distinct perturbative setups for which the convergence of the series expansion involved in the perturbation theory can be rigorously controlled. When the deformation is placed near the potential's extremum, the resulting corrections to the bound-state energies can be analytically continued to yield perturbed quasinormal frequencies, in agreement with known semi-analytic results. In contrast, when the perturbation is localized asymptotically far from the compact object, the bound states are only mildly modified and are accurately described by a perturbative expansion to the first order. However, the associated analytic continuation yields a strongly deformed spectrum that shows no clear connection to the quasinormal modes. These findings contribute to the effort to scrutinize the conditions under which bound states faithfully encode quasinormal spectra and to shed light on the underlying physics of black hole spectral instability.

gr-qc↗

arXiv2026

Aspire: Practical Batch Queries on Dynamic Encrypted Database with Full Leakage Suppression

Leakage-abuse attacks against encrypted database systems exploit semantic leakage and pose a severe threat to the confidentiality of plaintext data and queries. To address this problem, we propose a zero-leakage encrypted database that rules out leakage-abuse attacks, which fundamentally rely on semantic leakage. We first distinguish between two kinds of zero-leakage security. Ideal zero-leakage forbids the disclosure of any information beyond public parameters, whereas restricted zero-leakage additionally permits the leakage of metadata (such as database size and query timing). We show that restricted zero-leakage is sufficient to thwart leakage-abuse attacks and is achievable in practice. We then formally define adaptive restricted zero-leakage security and present Aspire, an encrypted database that satisfies this notion. Aspire integrates probabilistic batch code and minimal perfect hashing to support efficient batch keyword queries and dynamic updates while revealing only non-semantic metadata. Experimental results demonstrate the practicality and efficiency of Aspire, showing that Aspire can achieve an amortized query latency of less than 1.38 ms (7.53 s for a batch of 5461 keyword queries), with an amortized communication cost of approximately 2.95 KB (16.1 MB in total), on a $2^{22}$ key-value pairs database.

cs.CR↗

arXiv2026

Tensor triangular geometry of band algebras

In this paper, we study the tensor triangular geometry of band algebras $RB$, where $B$ is a finite left regular band and $R$ is a commutative Noetherian ring. We show that the Balmer spectrum of $\Perf(RB)$ is homeomorphic to $\Spec R\times L_B$, where $L_B$ is the support lattice of $B$. We also show that there is a bijection between the localizing ideals of $D(RB)$ and the stable subsets of $\Spec R\times L_B$. As a consequence, the telescope conjecture holds for $D(RB)$. Finally, we construct a band $B$ such that the generalized Nerves of Steel Conjecture fails for the non-rigid tensor triangulated category $\Perf(RB)$.

math.CT↗

arXiv2026

MAGIC: Learning from Visibility Asymmetry for Unsupervised Stereo Matching

Learning disparity in occluded regions remains difficult for unsupervised stereo matching. Photometric supervision lacks valid target-view correspondences in these regions, while the teacher and student in conventional binocular self-training share the same target view and therefore the same occlusions. Even when supervision is available, the small proportion of occluded pixels limits their contribution to training. We propose MAGIC, a multi-baseline geometric consistency framework for reliable and effective occlusion supervision. The teacher and student share a reference image but use different target views, allowing the teacher to observe correspondences that are occluded from the student. After aligning disparities across baselines, MAGIC uses predictions from teacher-visible regions to supervise student-occluded regions. An occlusion-aware weighting strategy strengthens supervision on teacher-visible but student-occluded pixels, preventing their training signal from being overwhelmed by non-occluded regions. We also introduce MBS20K, a synthetic multi-baseline stereo dataset spanning diverse scenes, weather, and lighting. Pre-trained on MBS20K, MAGIC generalizes to real-world datasets with consistently fewer occluded-region outliers. On KITTI, the pre-trained model already outperforms several fine-tuned unsupervised methods. Fine-tuning this model on standard binocular pairs achieves state-of-the-art unsupervised performance on KITTI 2015 and 2012. Incorporating synthesized multi-baseline views during fine-tuning further improves performance. Our code and dataset will be released upon acceptance.

cs.CV↗

arXiv2026

PredActor: Predictive Action Diffusion for Steerable Onboard Humanoid Control

Diffusion models provide a flexible framework for motion generation, but turning this flexibility into closed-loop humanoid control remains challenging. Hierarchical generator-tracker systems steer motion through reference trajectories, yet these references may exceed the capabilities of the downstream tracker, leaving physical feasibility and disturbance recovery largely to a separate control module. Action-only diffusion avoids this separation by directly generating executable actions, but provides no explicit future-state trajectory that can be steered toward test-time motion objectives. Joint state-action diffusion offers a natural alternative, but existing controllers often rely on privileged full-body states, while learned behavior selection and test-time motion steering remain only partially integrated. We present PredActor, a predictive action diffusion policy that unifies both steering modes in one directly executed policy using proprioception alone. Given proprioceptive history and optional task context, PredActor jointly predicts actions and an internal future-state trajectory that enables guidance: classifier-free guidance strengthens text-conditioned motion, while classifier guidance steers future states toward test-time objectives. Only actions are executed, requiring neither a motion-reference tracker nor privileged full-body states. In simulation, PredActor reaches 44 of 45 destination targets and achieves a text retrieval score of 0.539 versus 0.424 for conditional action diffusion, with similar disturbance survival. Rolling denoising and computation-preserving runtime optimizations reduce the complete callback to 16.790 ms median and 19.383 ms p95 on a Jetson Orin NX, within the 20 ms control period. Deployed on a Unitree G1, PredActor demonstrates text-conditioned motion, disturbance response, joystick control, and semantic interpolation in simulation and hardware.

cs.RO↗

arXiv2026

Einstein--Langevin source correlations in closed non-relativistic systems

Stochastic semiclassical gravity represents stress--energy fluctuations through the Einstein--Langevin noise kernel. We examine its Newtonian projection for specified closed, particle-number-conserving non-relativistic sources and distinguish the source covariance from a reduced matter channel. The connected one-particle covariance is state dependent and differs from the universal potential-noise law postulated in the Diósi--Penrose model. To relate the source spectrum to dephasing, we define a separate static probe subject to a prescribed Gaussian noise law. For a finite-variance source operator in an isolated non-degenerate eigenstate, the full connected spectral measure has a gap around zero. The probe dephasing exponent is then bounded and its long-time slope vanishes. Stationary doublets can instead carry static or low-frequency spectral lines, while a specified freely spreading Gaussian source gives a finite limiting exponent. Continuum results depend on preparation, normalisation, infrared regularisation and the order of limits. We also compare coherent and incoherent many-body preparations and establish no-signalling for fixed classical-noise channels. These results identify the spatial, temporal and dynamical assumptions needed to relate source fluctuations to a matter-decoherence model. They do not exclude Markovian limits in other number-conserving non-relativistic preparations.

quant-ph↗

arXiv2026

GEM-KMeans: Memory-Efficient and Accurate Clustering on Massive Scale with GPU Optimization

Memory-efficient scaling on clustering problems without sacrificing statistical accuracy is of central interest for large-scale data analysis and machine learning problems. Nonnegative low-rank (NLR) matrix factorization for $K$-means is a scalable clustering method, which connects to semidefinite relaxations with optimal average-case exact recovery guarantees. However, a direct GPU implementation of NLR requires multiple large factor-sized buffers and substantial data movements that are essentially memory-bound. In this paper, we introduce GEM-KMeans, a spectrally normalized yet mathematically equivalent NLR formulation that fuses the gradient update, nonnegative projection, and sufficient statistics for normalization and iterate movement into a matrix-multiplication epilogue. Instead of retaining three massive factor-sized arrays, our IO-aware GPU implementation materializes only one single factor with small tile-reduction arrays as additional storage in the High Bandwidth Memory (HBM). We derive explicit memory costs and spectrally normalized smoothness bounds for optimizing the clustering objective function. Accurate clustering is demonstrated at massive scales on synthetic and real datasets, where performance gains of GEM-KMeans over existing GPU-accelerated Lloyd's algorithms involve data-dependent runtime tradeoffs.

cs.LG↗

arXiv2026

PanoFuse: Panorama-Enhanced Vision-Language-Action Learning with Decoupled Semantic-Geometric Routing

Vision-Language-Action (VLA) policies have shown promising performance in language-conditioned robotic manipulation. However, most existing VLA systems rely on conventional perspective cameras with limited fields of view, often missing global scene context and leading to unreliable manipulation under visual occlusions, distractors, and unseen environments. In this work, we propose PanoFuse, a panorama-enhanced VLA framework that complements local manipulation observations with global panoramic perception. PanoFuse introduces a dedicated panoramic branch that leverages a pretrained panoramic foundation model to extract complementary semantic and geometric representations from omnidirectional observations. Rather than directly mixing these heterogeneous features, we introduce Decoupled Semantic-Geometric Routing (DSGR), which maintains semantic and geometric representations as separate context streams and selectively routes both to downstream state and action representations through structured block-wise attention. This design provides the action expert with global spatial context while preserving task-relevant semantic information from the pretrained VLA backbone. We further develop a synchronized data collection pipeline and construct a new real-world manipulation dataset containing panoramic RGB observations, wrist-view images, language instructions, robot states, and actions. Across seven evaluation settings, PanoFuse achieves an average success rate of 52.9%, outperforming the evaluated baselines and achieving consistent gains under novel-object, unseen-background, and distractor-rich settings. Code and data will be released publicly at https://xux-hnu.github.io/PanoFuse.

cs.CV↗

arXiv2026

PSR: Predictive Sensorimotor Representation Learning for Contact-Rich Manipulation

Contact-rich manipulation requires policies to generate precise actions by reasoning over contact forces, robot configurations, and interaction histories beyond visual observations. Existing methods passively condition on force feedback rather than actively predicting future contact dynamics, limiting their ability to generate high-precision actions. To address this problem, we introduce Predictive Sensorimotor Representation (PSR) learning, a framework that learns a hierarchy of predictive representations from multimodal sensorimotor signals and integrates them into the action stream of a visuomotor policy. Specifically, during a pretraining stage, a multimodal Transformer is trained to learn a hierarchy of predictive representations by jointly forecasting future interaction dynamics. The learned hierarchy subsequently augments the action stream, enabling the resulting policy to exploit contact-relevant cues at multiple depths. We further instantiate PSR within a Vision-Language-Action (VLA) model, resulting in PSR-VLA, and evaluate it on six real-world contact-rich manipulation tasks. Experimental results show that PSR-VLA achieves 91.7% overall success, improving over $π_{0.5}$, ForceVLA-$π_{0.5}$, and ForceVLA2-$π_{0.5}$ by 30.0, 22.5, and 19.2 percentage points, respectively. These results demonstrate the effectiveness of the proposed PSR for force-aware, contact-rich manipulation. Videos of the tasks and stability tests are available at https://psr-vla.pages.dev/.

cs.RO↗

arXiv2026

Correlated Galactic Confusion Foreground for the LISA-Taiji-TianQin Network and Its Impacts on Resolvable-Source Analysis

The millihertz gravitational-wave sky is expected to be observed by a network of space-based detectors (LISA, Taiji, and TianQin) in the 2030s. A dominant noise in this band is the confusion foreground produced by O(10^7) unresolved Galactic binaries (GBs). Because the same population projects onto the time-delay interferometry (TDI) channels of the different detectors, the foreground is necessarily correlated across the network. We construct the full foreground noise covariance matrix for the LISA-Taiji-TianQin network from a catalogue of ~3 x 10^7 GBs through numerical simulation, and we derive an analytic model of the cross-detector foreground coherence that provides a cross-check and physical interpretation of the numerical results. We derive the overall sensitivity of the detector network based on this covariance matrix, and further characterize the frequency- and time-dependent cross-detector foreground correlations. Taking massive black hole binaries (MBHBs) and GBs as representative transient and continuous sources, we further compare the block-diagonal (i.e., neglecting the cross-detector foreground correlation) and full-covariance noise models in terms of their impacts on the signal-to-noise ratio (SNR), parameter uncertainties, and Bayesian posteriors. The impact is confined to specific signal regimes, affecting primarily high-mass (M_c ~ 10^7 M_sun) MBHBs and low-frequency (f_0 <~ 2 mHz) GBs, with SNR relative differences of up to ~30% and ~10%, respectively. No statistically significant parameter bias arises under either noise model, as verified by probability-probability tests. Parameter estimations for MBHBs and GBs are performed using the network analysis pipelines implemented in the Triangle-BBH and Triangle-GB codes. Both repositories, together with the simulated foreground data and network sensitivities, are publicly released for diverse scientific investigations.

gr-qc↗

arXiv2026

Fast CZ gate in hybrid fluxonium-transmon systems with tunable couplers

Hybrid superconducting architectures combining different types of qubits offer a promising platform for exploiting their complementary advantages, yet high-fidelity entangling gates remain challenging because of strong nonlinearities and residual qubit-qubit interactions. Here, we propose a high-fidelity controlled-Z (CZ) gate for a hybrid circuit comprising a fluxonium qubit, a fixed-frequency transmon qubit, and a flux-tunable transmon coupler. By modulating only the external magnetic flux applied to the coupler, the qubit-qubit interaction is dynamically engineered for conditional-phase accumulation while the residual interaction is suppressed at idle, mitigating spectator-induced errors. We employ a low-dimensional Fourier-cosine pulse parameterization and a physically motivated cost function to independently suppress conditional-phase errors and leakage from the computational subspace. Numerical simulations demonstrate that a microwave-free CZ gate can be realized within $25 \mathrm{ns}$, with an average gate fidelity exceeding $99.99\%$ and leakage below $10^{-5}$. Using experimentally relevant superconducting-qubit parameters and accounting for decoherence, the proposed scheme maintains a CZ-gate fidelity of approximately $99.9\%$. We further extend the analysis to larger coupled architectures and find that the CZ-gate infidelity remains below $10^{-4}$ in the presence of spectator qubits. These results establish single-parameter flux control as a simple and robust approach for realizing high-fidelity entangling gates in heterogeneous superconducting quantum architectures.

quant-ph↗

arXiv2026

Self-supervised reconstruction of transients in data from space-borne gravitational-wave detectors

Space-based gravitational-wave (GW) data may contain transient signals whose waveform morphologies are not known in advance. Extracting these signals is important for characterizing new sources and mitigating instrumental anomalies. However, existing deep neural network (DNN)-based extraction approaches rely on clean training targets and waveform-class-specific examples, which may limit their applicability when the transient morphology is not specified in advance. This work develops a Noise2Noise (N2N)-inspired self-supervised framework that learns from noisy observations without clean training targets and requires no transient-specific waveform templates at inference. A single model trained on noisy massive black-hole binary (MBHB) observations provides high-overlap MBHB recovery and recovers the dominant morphologies of instrumental glitches and other GW transient signals such as cosmic string bursts in source-confused test data. Beyond waveform recovery, when independent information identifies a candidate transient as instrumental, its extracted waveform can be subtracted from the data without an anomaly-specific template. In a simulated continuous data stream, this procedure substantially suppresses the injected-anomaly power within the glitch-dominated frequency band. These results support the use of self-supervised extraction for initial waveform estimation of candidate transients with unknown morphologies, enabling subsequent characterization and, where appropriate, conditional subtraction of instrumental anomalies.

gr-qc↗

arXiv2026

Stratification of Artin motives over local fields

Let k be a field of characteristic p>0. We prove that for every p-decomposition group P, the tensor-triangulated category \operatorname{DPerm}(P;k) is stratified and its Balmer spectrum is generically noetherian. As an arithmetic application, we show that for a nonarchimedean local field F with residue characteristic \ell, the category $\operatorname{DAM}(F;k)$ of derived Artin motives is stratified if and only if \ell\neq p. In the stratified case its Balmer spectrum is generically noetherian; consequently, \operatorname{DAM}(F;k) satisfies the telescope conjecture.

math.CT↗

arXiv2026

Git4Data: Database-Native Version Control for AI Agents

Large Language Model (LLM) agents increasingly explore many candidate states of relational data in parallel, each of which should remain isolated, reproducible, and auditable, preferably through the same SQL interface used for ordinary data work. Existing tools support this requirement only partially: source-code version control does not scale to large datasets, whereas relational databases manage large data efficiently but rarely expose native branching, comparison, and merging. We present Git4Data, a database-native version-control layer for agentic workflows. Git4Data treats a database as a repository and a table as a versioned object, exposing Git-style operations (snapshot/tag, branch, diff, and merge with explicit conflict-resolution policies) through SQL extensions. Implemented in MatrixOne, a cloud-native relational database, Git4Data leverages immutable object storage and MVCC to make the cost of these operations proportional to the size of the change rather than the size of the data. On the BranchBench agentic branching workloads, Git4Data outperforms DoltDB by up to an order of magnitude. Overall, we believe this work sheds light on how relational databases can better support AI agents through efficient versioning.

cs.DB↗

arXiv2026

ContextPipe: Database-Inspired Context Assembly for Long-Horizon Agents

Long-horizon large language model (LLM) agents require context assembly: the runtime must decide what to include in each prompt, in what order, and when to compact history under a hard context-window budget and a byte-sensitive prompt cache. In production agentic systems, this logic is scattered across prompt builders, ad hoc compaction routines, cache-break workarounds, and per-provider shims. We argue that context assembly is structurally isomorphic to query execution in a relational database: both execute under a hard budget, exploit a tiered cache, and leverage statistics. We adopt this discipline in ContextPipe: a five-phase pipeline (Plan Bind Optimize Execute Feedback) backed by a structured data-source catalog, a deterministic cache-aware optimizer, and an EXPLAIN ANALYZE trace. We show that context in ContextPipe is auditable, replayable, and failure-isolated. A preliminary evaluation using the SWE-bench Pro Qutebrowser subset shows that, compared with the append-only context construction policy, ContextPipe reduces total token volume by 31%, LLM calls by 23%, and response time by 9%, at the cost of a lower KV cache-hit ratio.

cs.AI↗

arXiv2026

Accelerating Stochastic Gravitational Wave Backgrounds Parameter Estimation in Pulsar Timing Arrays with Flow Matching

\Acp{PTA} are essential tools for detecting the \ac{SGWB}, but their analysis faces significant computational challenges. Traditional methods like \ac{MCMC} struggle with high-dimensional parameter spaces where noise parameters often dominate, % while existing deep learning approaches fail to model the \ac{HD} correlation or are validated only on synthetic datasets. while existing deep learning approaches have so far been validated on synthetic datasets or require training on the full pulsar set, incurring substantial computational and memory costs. We propose a flow-matching-based \ac{CNF} for efficient \ac{PTA} parameter estimation. Using ten pulsars selected according to published NANOGrav 12.5-year dropout factors and applied to the NANOGrav 15-year residuals, our method produces \ac{SGWB} posteriors consistent with a reference \ac{MCMC} analysis, with Jensen-Shannon divergences below $10^{-2}$ nat. After amortized training, posterior generation is reduced from approximately 50 hours for the reference \ac{MCMC} pipeline to approximately 4 minutes for \ac{CNF} sampling. The present study demonstrates that flow-matching-based \acp{CNF} can serve as an efficient posterior-sampling accelerator for reduced \ac{PTA} datasets. Rather than providing new astrophysical constraints, the method is intended to complement conventional \ac{MCMC} analyses and to provide a scalable route toward faster inference in future \ac{PTA} applications.

astro-ph.IM↗