arXiv subjects

Xin Zhang

Publications and source records attributed to Xin Zhang.

At least 181 records · Page 10Linked to original sources

arXiv2026

Modulational stability of the periodic traveling wave in a local model for shallow water waves

In this paper, we investigate the modulational stability of periodic traveling waves in a local model for shallow water waves, which is an extended version of the Hunter-Saxton equation. We construct a family of small-amplitude periodic traveling waves for this local model and provide a parameterization of these waves. Using Floquet-Bloch theory, perturbation theory, and spectral analysis, we then establish the modulational stability of these background periodic traveling wave solutions. Finally, we analyze the modulational instability of another extended Hunter-Saxton equation with cubic nonlinearities, following a similar approach.

math.AP↗

arXiv2026

Fundamental Physics and Cosmology with TianQin

The exploration of the surrounding world and the universe is an important theme in the legacy of humankind. The detection of gravitational waves is adding a new dimension to this grand effort. What are the fundamental physical laws governing the dynamics of the universe? What is the fundamental composition of the universe? How has the universe evolved in the past and how will it evolve in the future? These are the basic questions that press for answers. The space-based gravitational wave detector TianQin will tune in to gravitational waves in the millihertz frequency range ($10^{-4} \sim 1$ Hz, to be specific), opening a new gravitational wave spectrum window to explore many of the previously hidden sectors of the universe. TianQin will discover many astrophysical systems, populating the universe at different redshifts: some will be of new types that have never been detected before, some will have very high signal-to-noise ratios, and some will have very high parameter estimation precision. The plethora of information collected will bring us to new fronts on which to search for the breaking points of general relativity, the possible violation of established physical laws, the signature of possible new gravitational physics and new fundamental fields, and to improve our knowledge on the expansion history of the universe. In this white paper, we highlight the advances that TianQin can bring to fundamental physics and cosmology.

gr-qc↗

arXiv2026

A Wireless Reconfigurable Metasurface for Enhanced Parallel Magnetic Resonance Imaging

Modern magnetic resonance imaging (MRI) relies on application-specific multi-channel receive coils to achieve high performance, but these coils are typically costly, rigid, and difficult to generalize across anatomies. Recent wireless, low-cost metamaterials offer improved signal-to-noise ratio (SNR) but remain anatomy-dependent, are prone to destructive inter-element interference, and lack demonstrated compatibility with parallel imaging. Herein, a wireless, reconfigurable coaxial loop metasurface (CLM) is introduced as a platform for localized SNR enhancement that can operate either as a standalone element or as an insertable add-on alongside existing clinical receive systems. Through its coaxial architecture and shared current pathways, the CLM establishes a collective in-phase resonant mode that enforces phase-coherent current distributions across all loops, resulting in consistently constructive interference. Benchmarking on a 3.0 T MR system using an 8-loop CLM shows SNR enhancements of up to 14.8-fold and 14.02-fold in the sagittal and axial planes, relative to the birdcage coil (BC). As an add-on to a clinical posterior receive array, it further demonstrates up to 2.9-fold SNR enhancement and compatibility with parallel imaging across ex vivo and in vivo settings. The proposed CLM paves the way toward a new class of reconfigurable and insertable MRI hardware for flexible and system-compatible signal enhancement.

physics.med-ph↗

arXiv2026

NeurIPS: Neuro-anatomical Inductive Priors for Sphere-based Brain Decoding

Current fMRI decoders face a performance-fidelity trade-off where efficient ID encoders outperform geometrically faithful surface-based models. We argue this is partly driven by inefficient surface tokenization and the failure to use anatomy as a predictive signal. We present NeurIPS, a framework that improves surface-based decoding by reframing anatomical variation from a nuisance to a powerful inductive prior. NeurIPS unites two innovations: a Selective ROI Spherical Tokenizer (SRST) for efficient geometric encoding, and a Structure-Guided Mixture of Experts (SG-MoE) that explicitly models individual anatomy using cortical features. On the Natural Scenes Dataset, NeurIPS establishes a new state-of-the-art for surface decoders and achieves performance comparable to strong 1D baselines. This is achieved with unprecedented efficiency, as the model converges dramatically faster (10 vs. 600 epochs). This efficiency enables rapid adaptation to new subjects using only 20% of data and ensures robust scalability as the training cohort is expanded. Ablations provide causal evidence that these gains are driven by the model's use of cortical features, not by memorizing subject IDs. By leveraging anatomical priors, NeurIPS provides a principled and scalable path toward robust, generalizable brain decoding.

cs.AI↗

arXiv2026

SEAL: Synergistic Co-Evolution of Agents and Learning Environments

Large Language Model (LLM) agents are increasingly improved through interaction, yet most self-evolution methods adapt either the policy or the learning environment in isolation. We identify this structural gap as \emph{Agent-Environment Misalignment}: the agent's capability frontier changes during training, while the environment that provides supervision remains static or only weakly coupled to the agent's revealed failures. We propose SEAL, a closed-loop co-evolution framework for interactive tool-use agents. SEAL collects on-policy trajectories under executable verification, diagnoses failed rollouts into turn-level failure labels, and uses these diagnoses as a shared signal for both environment-side adaptation and model-side policy optimization. The environment evolves its training-time learning interface by exposing clearer tool affordance cues, constraint information, and recovery-oriented feedback, while the policy is updated with diagnosis-guided advantage reweighting. Extensive experiments across in-distribution and out-of-distribution multi-turn tool-use evaluations show that SEAL improves low-resource agent learning: with only 400 training samples, it yields +8.25 to +26.25 average-point gains across three backbones and exhibits positive out-of-distribution transfer. These results demonstrate the value of jointly adapting the learner and its training-time learning substrate for robust self-improving LLM agents.

cs.CL↗

arXiv2026

SpikingMoE: SDPrompt-Guided Dynamic Expert Fusion in Spiking Neural Networks

Spiking Neural Networks (SNNs) provide an energy-efficient paradigm for visual recognition. We present SpikingMoE, which integrates a spike-driven Transformer with a Mixture-of-Experts (MoE) framework for dynamic computation. Inspired by the lateral geniculate nucleus (LGN), a spike-driven prompt (SDprompt) enables input-dependent expert routing in a biologically plausible manner. By replacing standard MLPs with spike-compatible expert modules and enforcing binary spike communication, SpikingMoE is designed for neuromorphic hardware. Experiments on CIFAR-10 and CIFAR-100 achieve 94.09% and 74.54% top-1 accuracy, showing that modular expert routing can be incorporated while retaining reasonable performance. To our knowledge, SpikingMoE is the first open-source SNN framework that integrates MoE into a spike-driven Transformer with LGN-inspired routing.

cs.NE↗

arXiv2026

Generalized specific entropy on Wiener space with application to Martingale Optimal Transport

Classical entropy regularization is poorly suited to continuous-time martingale transport, since relative entropy between diffusion laws typically forces their volatility characteristics to coincide. We introduce a specific-entropy framework based on Poisson jump approximations of continuous martingales. In the Gaussian-mark case, this yields explicit generalized specific entropy functionals on Wiener space, whose limiting costs depend not only on the limiting martingale laws but also on the microscopic approximation mechanism. This Poissonization approach avoids deterministic grid refinement and the associated high-dimensional multimarginal Sinkhorn problems, while allowing jump intensities to reflect local volatility. We prove weak convergence of the Poisson approximations and identify the limiting entropy functionals. For a trace-normalized Poisson scheme, the resulting cost defines a continuous-time specific-entropic martingale optimal transport problem, called SEMOT. This cost yields compactness, existence, and strong duality, and leads formally to a coupled Hamilton-Jacobi-Bellman/Fokker-Planck system. The resulting structure suggests Sinkhorn type numerical schemes, which we implement in one and two dimensions.

math.PR↗

arXiv2026

Draw2Think: Harnessing Geometry Reasoning through Constraint Engine Interaction

Vision-language models solve geometry problems with rising accuracy, yet their intermediate states remain latent and unverifiable: a relation expressed in textual reasoning or drawing code carries no guarantee that a constraint-satisfying configuration realizes it. We observe that existing externalization methods based on rendered pixels or one-shot scripts fail to provide exact, per-action geometric guarantees. Enforcing geometric relations by algebraic definition closes this gap: the workspace becomes a constraint-checked evolving canvas. We present Draw2Think, a framework that recasts geometric reasoning from latent spatial inference into agentic interaction with the GeoGebra constraint engine. In a Propose-Draw-Verify loop, Draw2Think externalizes hypotheses onto an executable canvas, measures exact geometric quantities, and feeds structured observations back to the model, so subsequent reasoning proceeds from checked canvas state grounded by the shared workspace. This externalization makes two properties separately auditable: model-level Construction Fidelity (whether the canvas realizes the intended configuration) and engine-level Measurement Faithfulness (exact values and relations from canvas constraints). Across construction, outcome, and rendering evaluations, Draw2Think builds canvases that pass 95.9% predicate-level and 84.0% strict problem-level construction checks on GeoGoal, improves outcome accuracy by up to 4.1%/16.4% on planar/solid benchmarks, and attains 68.2%/90.5% strict/relaxed rendering scores on GenExam-math. Project page is available at https://draw2think.github.io/

cs.CV↗

arXiv2026

Contrastive self-supervised convolutional autoencoder for core-collapse supernova gravitational-wave detection

Gravitational-wave astronomy has opened a direct observational window onto compact-object dynamics, strong-field gravity, and cosmology. Among the transient sources accessible through this window, core-collapse supernovae (CCSNe) are uniquely valuable because their signals can probe the engine of stellar collapse, proto-neutron-star dynamics, and explosion asymmetries, yet their weak, stochastic, and model-dependent waveforms remain difficult to detect. In this work, we develop a contrastive self-supervised convolutional autoencoder (CS-CAE) for CCSNe gravitational-wave signal detection. The method combines a convolutional autoencoder (CAE), a noise-centered latent regularizer, and a projection head trained with a contrastive objective. This design encourages independent noisy realizations of the same CCSNe signal to be mapped to nearby latent representations, thereby reducing the influence of random noise fluctuations. CS-CAE achieves performance comparable to a supervised convolutional neural network while clearly outperforming a conventional CAE baseline, and generalizes better to unseen numerical CCSNe waveform families. Under the Einstein Telescope (ET) detector configuration, the method achieves an effective sensitive distance of approximately 120 kpc and shows improved separation of CCSNe signals from stationary noise and transient glitches in the low-false-alarm regime. These results highlight the potential of CS-CAE as a robust and less template-dependent framework for CCSNe gravitational-wave searches.

gr-qc↗

arXiv2026

WorldArena 2.0: Extending Embodied World Model Benchmarking on Modality, Functionality and Platform

World models have emerged as a central paradigm for embodied intelligence, enabling agents to predict action-conditioned future and reason about environmental dynamics. However, existing embodied world model benchmarks are still largely confined to vision-only prediction, offline embodied applications, and simulator-based evaluation, making them insufficient for assessing increasingly comprehensive world models. In this work, we introduce WorldArena 2.0, an expanded benchmark that systematically broadens embodied world model evaluation along three dimensions: modality, functionality, and platform. Along the modality dimension, WorldArena 2.0 extends evaluation from vision-only to visuotactile modalities, enabling assessment of multimodal perception and prediction. Along the functionality dimension, it extends beyond policy evaluation and planning to assess world models as interactive RL environments for policy optimization. Along the platform dimension, it moves beyond simulator-only evaluation to a diverse suite of simulated and real-world robotic settings across multiple embodiments. Under a standardized protocol, WorldArena 2.0 comprehensively evaluates perceptual quality, interactive utility, and cross-platform performance, providing a comprehensive testbed for tracking progress toward embodied world models. The benchmark is available at: https://world-arena.ai.

cs.RO↗

arXiv2026

Guiding LLM-based Loop Invariant Synthesis via Feedback on Local Reasoning Errors

We propose a novel framework that provides constructive feedback to an LLM in the "guess-and-check" paradigm by formally verifying its own thinking process and detecting local reasoning errors. We apply this framework to the loop invariant synthesis problem. We prompt the model to produce a step-by-step natural language proof justifying its thinking process for the failed verification condition of its generated loop invariants. Then, we use an LLM to translate the reasoning steps into first-order logic implications, which can be checked automatically. An invalid implication pinpoints the exact logical flaw in the LLM's thinking process, which we then use to construct targeted feedback for refinement. We have implemented our approach in a tool called LORIS and evaluated it on a main benchmark suite of 460 C programs and an additional benchmark suite of 50 C programs each of which involves non-linear properties. On the main benchmark suite, LORIS solved 445 of the programs, and achieved an overall success rate of $93.1\%$. LORIS also demonstrates robustness on the challenging non-linear benchmark suite.

cs.PL↗

arXiv2026

Stellar Density Classification and Regression for CSST Multi-color Imaging Using Deep Learning

The Chinese Space Station Survey Telescope (CSST) aims to map the universe across an unprecedented dynamic range of stellar densities, spanning from extragalactic voids to the crowded Galactic center (e.g. a few stars and galaxies in the voids and $>10^5$ stars per detector in Galactic center). However, processing such heterogeneous data with a general source extraction pipeline introduces significant systematic uncertainties, standard algorithms exhibit poor accuracy in crowded fields and suffer from increased astrometric uncertainty in void regions. To mitigate these systematics, we propose a hierarchical, two-stage deep learning model for adaptive data reduction. The first stage ('classification') employs a ResNet-34 model to classify images into six discrete density categories, achieving $98.83\%$ in global accuracy. This classification acts as a critical decision gate, ensuring high calibration accuracy in the crowded fields. In the second stage ('regression'), a ResNet-50 regression model predicts the bright stars ($<23.5$ mag) in the field, which is essential for astrometric calibration, achieving a mean absolute error (MAE) of 0.0824 dex. By decoupling density characterization from source extraction, our model ensures that photometric and astrometric algorithms are optimally matched to the stellar density environment, thereby enhancing the fidelity and homogeneity of CSST as well as future large sky survey data products.

astro-ph.IM↗

arXiv2026

Extracting redshifts from 2D slitless spectroscopic images using deep learning for the CSST galaxy survey

Wide-field slitless spectroscopic galaxy surveys, such as the one performed by the upcoming Chinese Space Station Survey Telescope (CSST), are crucial for precision cosmology but present formidable data analysis challenges. Because spectra are dispersed directly onto the detector, they are convolved with the 2-dimensional (2D) spatial morphology, which complicates wavelength calibration and consequently degrades the fidelity of subsequent 1-dimensional (1D) spectral extraction. To overcome these limitations, we present a deep learning framework that extracts redshifts directly from 2D slitless spectral images, bypassing 1D extraction entirely. We construct a realistic mock dataset for the CSST $GV$ and $GI$ band using high-resolution images from HSC-SSP PDR3 and spectral energy distributions (SEDs) from DESI DR1. A Bayesian convolutional neural network implemented by Monte Carlo dropout is employed to map the 2D spectral images to redshift estimations while simultaneously quantifying uncertainties. We find that our model can achieve a precision $σ_{\rm NMAD}=0.0104$ and mean uncertainty $\langle E / (1 + z_{\rm true}) \rangle=0.0155$ for sources with ${\rm SNR}_{GI}\geq1$. For sources with ${\rm SNR}_{GI}$ higher than 3.0, 5.0 and 10.0, $σ_{\rm NMAD}$ can achieve 0.0047, 0.0037 and 0.0024 respectively, matching the redshift precision requirements for studies such as BAO using the CSST slitless spectroscopic surveys. Furthermore, by utilizing spatial augmentations, the network demonstrates resilience to wavelength calibration errors. This work provides a novel and robust pathway for data analysis of next-generation slitless spectroscopic galaxy surveys.

astro-ph.IM↗

arXiv2026

Exact nonequilibrium steady states of boundary driven circuit with XYZ gates

We obtain the exact many-body density operator of a boundary-driven XXZ quantum circuit via a spatially inhomogeneous matrix product Ansatz. The Ansatz has formally infinite bond-dimension and generalizes authors' previous construction \cite{2025XXZcircuit} for the XXZ interactions. The boundary qubits are coupled to reset quantum channels that project them toward arbitrary pure target states. We find and describe a family of relatively robust separable chiral nonequilibrium steady states (NESS), which are elliptic analogs of spin helices for the circuit, and which are particularly attractive from an experimental perspective.

quant-ph↗

arXiv2026

WorldVLN: Autoregressive World Action Model for Aerial Vision-Language Navigation

Aerial vision-language navigation (VLN) requires agents to follow natural-language instructions through closed-loop perception and action in 3D environments. We argue that aerial VLN can be formulated as a prediction-driven world-action problem: the agent should anticipate latent world evolution and act according to the predicted consequences. To this end, we propose WorldVLN, the first autoregressive world action model for aerial VLN. Unlike full-sequence video-generation world models that generate an entire visual clip, WorldVLN adapts a latent autoregressive video backbone to predict short-horizon world-state transitions and directly decodes them into executable waypoint actions. After each action segment is executed, newly received observations are encoded back into the autoregressive context, enabling closed-loop world-action prediction. We further introduce a two-stage training framework that first grounds the video prior in instruction-conditioned navigation dynamics and then develops Action-aware GRPO, the first reinforcement learning method tailored to autoregressive WAMs, to optimize waypoint decisions through their downstream rollout consequences. On public outdoor and indoor benchmarks, WorldVLN consistently outperforms existing Vision-Language-Action baselines with 12\%+ success-rate gains and larger advantages on challenging cases. It further transfers zero-shot to real drone deployment, suggesting that the proposed WorldVLN offers a promising route for spatial action tasks. Demos and code are available at https://embodiedcity.github.io/WorldVLN/.

cs.RO↗

arXiv2026

Physics-Aware Machine-Learning-Driven Inverse Design of Broadband Ultra-Open Acoustic Metamaterials

Ventilated acoustic silencers combing sound attenuation with high ventilation are pivotal for advanced noise control. However, balancing attenuation, bandwidth, openness, and thickness remains a high-dimensional challenge. Here, we report a physics-aware machine-learning-driven inverse design framework for ultra-open acoustic silencers (UAS). By leveraging Green's function-based parameterization, we physically decouple the design space into spectral and radial parameters, ensuring physical interpretability while reducing complexity. We introduce a two-stage forward prediction architecture that captures broadband envelopes and sharp resonant features via a coarse-to-fine strategy. Coupled with a population-based, hybrid-objective parallel (PHP) inverse strategy, our framework enables rapid exploration of non-convex landscapes, identifying hundreds of optimized candidates within seconds. Crucially, this framework uncovers hidden linear design rules that govern high-performance monolithic designs, acting as geometric proxies for optimal impedance-matching. We experimentally validate a family of prototypes: UAS-2 demonstrates the monolithic limit with high ventilation ratio, while UAS-3 demonstrates versatility in multi-mode interactions. To circumvent the trade-off ceiling of single-unit resonators, a parallel-composite architecture (UAS-4) is introduced to enhance performance through spatial interference distribution. Results confirm a broadband bandwidth exceeding 830 Hz achieved with an ultra-thin profile (0.1-0.2λ) and 80% ventilation. This work establishes a data-driven paradigm for discovering design principles in functional metamaterials.

physics.app-ph↗

arXiv2026

SGR: A Stepwise Reasoning Framework for LLMs with External Subgraph Generation

Large Language Models (LLMs) have demonstrated strong capabilities across diverse NLP applications, such as translation, text generation, and question answering. Nevertheless, they remain limited in complex settings that demand deep reasoning and logical inference. Since these models are trained on large-scale text corpora, their generation process may still introduce irrelevant, noisy, or factually inconsistent content. To mitigate this problem, we introduce SGR, a stepwise framework that enhances LLM reasoning through external subgraph generation. SGR builds query-specific subgraphs from external knowledge bases and uses their semantic structure to support multi-step inference. By grounding intermediate reasoning steps in structured external knowledge, the framework helps the model concentrate on relevant entities, relations, and supporting evidence. In particular, SGR first constructs a subgraph tailored to the input question. It then guides the model to reason progressively over the generated structure and combines multiple reasoning trajectories to obtain the final prediction. Experimental results across several benchmark datasets show that SGR achieves consistent improvements over competitive baselines, highlighting its value for improving both reasoning accuracy and factual reliability.

cs.CL↗

arXiv2026

A Tale of Two Problems: Multi-Task Bilevel Learning Meets Equality Constrained Multi-Objective Optimization

In recent years, bilevel optimization (BLO) has attracted significant attention for its broad applications in machine learning. However, most existing works on BLO remain confined to the single-task setting and rely on the lower-level strong convexity assumption, which significantly restricts their applicability to modern machine learning problems of growing complexity. In this paper, we make the first attempt to extend BLO to the multi-task setting under a relaxed lower-level general convexity (LLGC) assumption. To this end, we reformulate the multi-task bilevel learning (MTBL) problem with LLGC into an equality constrained multi-objective optimization (ECMO) problem. However, ECMO itself is a new problem that has not yet been studied in the literature. To address this gap, we first establish a new Karush-Kuhn-Tucker (KKT)-based Pareto stationarity as the convergence criterion for ECMO algorithm design. Based on this foundation, we propose a weighted Chebyshev (WC)-penalty algorithm that achieves a finite-time convergence rate of $O(ST^{-\frac{1}{2})$ to KKT-based Pareto stationarity in both deterministic and stochastic settings, where $S$ denotes the number of objectives, and $T$ is the total iterations. Moreover, by varying the preference vector over the $S$-dimensional simplex, our WC-penalty method systematically explores the Pareto front. Finally, solutions to the ECMO problem translate directly into solutions for the original MTBL problem, thereby closing the loop between these two foundational optimization frameworks.

cs.LG↗