arXiv subjects

Shu Zhang

Publications and source records attributed to Shu Zhang.

At least 37 records · Page 2Linked to original sources

arXiv2026

Boundary Floquet Control of Bulk non-Hermitian Systems

Boundary perturbations are generally irrelevant for bulk properties in the thermodynamic limit, as they are edge-confined and subextensive. We show that this expectation breaks down in boundary-driven systems exhibiting the non-Hermitian skin effect, where arbitrarily weak boundary Floquet driving reconstructs bulk quasienergy spectra and dynamics. We develop a Floquet non-Bloch band theory that extends generalized Brillouin-zone methods to boundary-driven systems at arbitrary driving frequencies, overcoming the lack of a general framework beyond high-frequency approximations. With representative single- and two-band models, we demonstrate that the boundary driving frequency tunes non-Bloch parity-time symmetry breaking, while its amplitude acts as a finite-size control parameter. Our work establishes boundary Floquet control as a general route for manipulating bulk properties, opening a new avenue for dynamical engineering in driven open systems.

quant-ph↗

arXiv2026

Precise $^{136}$Xe Double Beta Decay Measurement in PandaX-4T with Implications on the Nuclear Matrix Elements and Majorons

The continuous spectrum of double beta decay ($ββ$) provides a sensitive probe to test the predictions of the standard model and to search for signatures of new physics beyond it. We present a comprehensive analysis of the $^{136}$Xe $ββ$ spectrum utilizing $39.1 \pm 0.7~\textrm{kg}\cdot\textrm{yr}$ of $^{136}$Xe exposure from the PandaX-4T experiment. The analysis yields the most precise measurement to date of the $^{136}$Xe two-neutrino double beta decay ($2νββ$) half-life, $(2.14 \pm 0.05) \times 10^{21}$ years, the uncertainty of which is reduced by a factor of 2 compared to our previous result. We measure the parameter $ξ_{31}^{2ν}$, defined as the ratio between the subleading and leading components of the $^{136}$Xe $2νββ$ nuclear matrix element, to be $0.59^{+0.41}_{-0.38}$, which is consistent with theoretical predictions. We also search for Majoron-emitting modes of $^{136}$Xe $ββ$, establishing the most stringent limit for the spectral index $n=7$.

nucl-ex↗

arXiv2026

Spin-qubit Noise Spectroscopy of Magnetic Berezinskii-Kosterlitz-Thouless Physics

We propose using spin-qubit noise magnetometry to probe dynamical signatures of magnetic Berezinskii-Kosterlitz-Thouless (BKT) physics. For a nitrogen-vacancy (NV) center coupled to two-dimensional XY magnets, we predict distinctive features in the magnetic noise spectral density in the sub-MHz to GHz frequency range. In the quasi-long-range ordered phase, the spectrum exhibits a temperature-dependent power law characteristic of algebraic spin correlations. Above the transition, the noise reflects the proliferation of free vortices and enables quantitative extraction of the vortex conductivity, a key parameter of vortex transport. These results highlight NV as a powerful spectroscopic method to resolve magnetic dynamics in the mesoscopic and low-frequency regimes and to probe exotic magnetic phase transitions.

cond-mat.mes-hall↗

arXiv2026

ESL-Bench: An Event-Driven Synthetic Longitudinal Benchmark for Health Agents

Longitudinal health agents must reason across multi-source trajectories that combine continuous device streams, sparse clinical exams, and episodic life events - yet evaluating them is hard: real-world data cannot be released at scale, and temporally grounded attribution questions seldom admit definitive answers without structured ground truth. We present ESL-Bench, an event-driven synthesis framework and benchmark providing 100 synthetic users, each with a 1-5 year trajectory comprising a health profile, a multi-phase narrative plan, daily device measurements, periodic exam records, and an event log with explicit per-indicator impact parameters. Each indicator follows a baseline stochastic process driven by discrete events with sigmoid-onset, exponential-decay kernels under saturation and projection constraints; a hybrid pipeline delegates sparse semantic artifacts to LLM-based planning and dense indicator dynamics to algorithmic simulation with hard physiological bounds. Users are each paired with 100 evaluation queries across five dimensions - Lookup, Trend, Comparison, Anomaly, Explanation - stratified into Easy, Medium, and Hard tiers, with all ground-truth answers programmatically computable from the recorded event-indicator relationships. Evaluating 13 methods spanning LLMs with tools, DB-native agents, and memory-augmented RAG, we find that DB agents (48-58%) substantially outperform memory RAG baselines (30-38%), with the gap concentrated on Comparison and Explanation queries where multi-hop reasoning and evidence attribution are required.

cs.AI↗

arXiv2026

Sign control of photocurrents by spin-group-symmetry breaking in altermagnetic insulators

Controlling physical responses through symmetry breaking is a central paradigm in quantum materials, enabling novel functionalities. Here we determine the effects of spin-group-symmetry breaking on nonlinear optical responses of collinear altermagnetic insulators. Using shear strain as an example, we show that the direction of symmetry-breaking induced components of charge and spin photocurrents are locked to the sign of the strain. In the absence of spin-orbit coupling, this effect is intuitively captured by the spin-gap asymmetry--an imbalance between spin-up and spin-down direct band gaps which couples trilinearly with the Néel order and the strain. We demonstrate this mechanism with density functional theory calculations on the recently proposed altermagnet CuWP$_2$S$_6$. Having symmetry-guided control of both charge and spin photocurrents allows, vice versa, to reveal and investigate altermagnetism in insulating materials by exploration of their optical responses.

cond-mat.mes-hall↗

arXiv2026

Structured Diversity Control: A Dual-Level Framework for Group-Aware Multi-Agent Coordination

Controlling the behavioral diversity is a pivotal challenge in multi-agent reinforcement learning (MARL), particularly in complex collaborative scenarios. While existing methods attempt to regulate behavioral diversity by directly differentiating across all agents, they lack deep characterization and learning of multi-agent composition structures. This limitation leads to suboptimal performance or coordination failures when facing more complex or challenging tasks. To bridge this gap, we introduce Structured Diversity Control (SDC), a framework that redefines the system-wide diversity metric as a weighted combination of intra-group diversity, which is minimized for cohesion and inter-group diversity, which is maximized for specialization. The trade-off is governed by a pre-set Diversity Structure Factor (DSF), allowing for fine-grained, group-aware control over the collective strategy. Our method directly constrains the policy architecture without altering reward functions. This structural definition of diversity enables SDC to deliver substantial performance gains across various experiments, including increasing average rewards by up to 47.1\% in multi-target pursuit and reducing episode lengths by 12.82\% in complex neutralization scenarios. The proposed method offers a novel analytical perspective on the problem of cooperation in group-aware multi-agent systems.

cs.AI↗

arXiv2026

Fake It Right: Injecting Anatomical Logic into Synthetic Supervised Pre-training for Medical Segmentation

Vision Transformers (ViTs) excel in 3D medical segmentation but require massive annotated datasets. While Self-Supervised Learning (SSL) mitigates this using unlabeled data, it still faces strict privacy and logistical barriers. Formula-Driven Supervised Learning (FDSL) offers a privacy-preserving alternative by pre-training on synthetic mathematical primitives. However, a critical semantic gap limits its efficacy: generic shapes lack the morphological fidelity, fixed spatial layouts, and inter-organ relationships of real anatomy, preventing models from learning essential global structural priors. To bridge this gap, we propose an Anatomy-Informed Synthetic Supervised Pre-training framework unifying FDSL's infinite scalability with anatomical realism. We replace basic primitives with a lightweight shape bank with de-identified, label-only segmentation masks from 5 subjects. Furthermore, we introduce a structure-aware sequential placement strategy to govern the patch synthesis process. Instead of random placement, we enforce physiological plausibility using spatial anchors for correct localization and a topological graph to manage inter-organ interactions (e.g., preventing impossible overlaps). Extensive experiments on BTCV and MSD datasets demonstrate that our method significantly outperforms state-of-the-art FDSL baselines and SSL methods by 1.74\% and up to 1.66\%, while exhibiting a robust scaling effect where performance improves with increased synthetic data volume. This provides a data-efficient, privacy-compliant solution for medical segmentation. The code will be made publicly available upon acceptance.

cs.CV↗

arXiv2026

The Texture-Shape Dilemma: Boundary-Safe Synthetic Generation for 3D Medical Transformers

Vision Transformers (ViTs) have revolutionized medical image analysis, yet their data-hungry nature clashes with the scarcity and privacy constraints of clinical archives. Formula-Driven Supervised Learning (FDSL) has emerged as a promising solution to this bottleneck, synthesizing infinite annotated samples from mathematical formulas without utilizing real patient data. However, existing FDSL paradigms rely on simple geometric shapes with homogeneous intensities, creating a substantial gap by neglecting tissue textures and noise patterns inherent in modalities like CT and MRI. In this paper, we identify a critical optimization conflict termed boundary aliasing: when high-frequency synthetic textures are naively added, they corrupt the image gradient signals necessary for learning structural boundaries, causing the model to fail in delineating real anatomical margins. To bridge this gap, we propose a novel Physics-inspired Spatially-Decoupled Synthesis framework. Our approach orthogonalizes the synthesis process: it first constructs a gradient-shielded buffer zone based on boundary distance to ensure stable shape learning, and subsequently injects physics-driven spectral textures into the object core. This design effectively reconciles robust shape representation learning with invariance to acquisition noise. Extensive experiments on the BTCV and MSD datasets demonstrate that our method significantly outperforms previous FDSL, as well as SSL methods trained on real-world medical datasets, by 1.43% on BTCV and up to 1.51% on MSD task, offering a scalable, annotation-free foundation for medical ViTs. The code will be made publicly available upon acceptance.

cs.CV↗

arXiv2026

Referring Layer Decomposition

Precise, object-aware control over visual content is essential for advanced image editing and compositional generation. Yet, most existing approaches operate on entire images holistically, limiting the ability to isolate and manipulate individual scene elements. In contrast, layered representations, where scenes are explicitly separated into objects, environmental context, and visual effects, provide a more intuitive and structured framework for interpreting and editing visual content. To bridge this gap and enable both compositional understanding and controllable editing, we introduce the Referring Layer Decomposition (RLD) task, which predicts complete RGBA layers from a single RGB image, conditioned on flexible user prompts, such as spatial inputs (e.g., points, boxes, masks), natural language descriptions, or combinations thereof. At the core is the RefLade, a large-scale dataset comprising 1.11M image-layer-prompt triplets produced by our scalable data engine, along with 100K manually curated, high-fidelity layers. Coupled with a perceptually grounded, human-preference-aligned automatic evaluation protocol, RefLade establishes RLD as a well-defined and benchmarkable research task. Building on this foundation, we present RefLayer, a simple baseline designed for prompt-conditioned layer decomposition, achieving high visual fidelity and semantic alignment. Extensive experiments show our approach enables effective training, reliable evaluation, and high-quality image decomposition, while exhibiting strong zero-shot generalization capabilities.

cs.CV↗

arXiv2026

DRACO: a Cross-Domain Benchmark for Deep Research Accuracy, Completeness, and Objectivity

We present DRACO (Deep Research Accuracy, Completeness, and Objectivity), a benchmark of complex deep research tasks. These tasks, which span 10 domains and draw on information sources from 40 countries, originate from anonymized real-world usage patterns within a large-scale deep research system. Tasks are sampled from a de-identified dataset of Perplexity Deep Research requests, then filtered and augmented to ensure that the tasks are anonymized, open-ended and complex, objectively evaluable, and representative of the broad scope of real-world deep research use cases. Outputs are graded against task-specific rubrics along four dimensions: factual accuracy (accuracy), breadth and depth of analysis (including completeness), presentation quality (including objectivity), and citation quality. DRACO is publicly available at https://hf.co/datasets/perplexity-ai/draco.

cs.LG↗

arXiv2026

Evolution of a Long-Lived Deep-Seated Main-Sequence Magnetic Field During White Dwarf Cooling

We study the evolution of white dwarf (WD) magnetic fields that originate from core-convective dynamos during the main-sequence. Using stellar evolution and WD cooling models combined with magnetic field diffusion calculations, we demonstrate that a surviving field from the main-sequence can account for various features observed in magnetic WDs. In particular, the earlier emergence of stronger magnetic fields in more massive WDs, compared to older, less massive, and less magnetic ones, can be explained by this framework. This is because the magnetic boundary at the onset of WD cooling lies deeper in less massive WDs, resulting in a slower and weaker evolution of the surface magnetic field due to increasing electrical conductivity over time. We further show that many of the magnetic field strengths observed across different WD samples can be reproduced if the deep-seated field generated during the main sequence is comparable to predictions from magnetohydrodynamic simulations of core-convective dynamos, or if equipartition provides a valid scaling for the main-sequence dynamo. Additionally, our predictions for surface magnetic fields vary by a factor of 2 to 4 when higher-order modes of poloidal magnetic field expansion and turbulent diffusion driven by crystallization-induced convection are included. These effects should therefore be considered when investigating the origin of magnetic fields in individual WDs.

astro-ph.SR↗

arXiv2026

LDRNet: Large Deformation Registration Model for Chest CT Registration

Most of the deep learning based medical image registration algorithms focus on brain image registration tasks.Compared with brain registration, the chest CT registration has larger deformation, more complex background and region over-lap. In this paper, we propose a fast unsupervised deep learning method, LDRNet, for large deformation image registration of chest CT images. We first predict a coarse resolution registration field, then refine it from coarse to fine. We propose two innovative technical components: 1) a refine block that is used to refine the registration field in different resolutions, 2) a rigid block that is used to learn transformation matrix from high-level features. We train and evaluate our model on the private dataset and public dataset SegTHOR. We compare our performance with state-of-the-art traditional registration methods as well as deep learning registration models VoxelMorph, RCN, and LapIRN. The results demonstrate that our model achieves state-of-the-art performance for large deformation images registration and is much faster.

cs.CV↗

arXiv2026

Discovery of a New Spectral Transition in Swift J0243.6+6124 in the Sub-Eddington Regime

We conduct a detailed spectral analysis of the Galactic ultraluminous X-ray pulsar Swift J0243.6+6124 in its sub-Eddington regime, using Insight-HXMT and NICER observations during multiple outbursts including the 2018 giant outburst. We discover a new transition at $L_{\rm t} \approx 4.5 \times 10^{37}\ {\rm erg\ s^{-1}}$, accompanied by systematic evolution of spectral parameters, in particular a significant turnover in the blackbody normalization. This transition luminosity in the sub-Eddington regime represents the fifth transition identified so far in Swift J0243.6+6124, further highlighting the complexity of its accretion-powered emission. We interpret the transition in terms of a multipolar magnetic-field configuration, where weak ($\sim 2.8 \times 10^{12}\ {\rm G}$) and strong ($\sim 1.6 \times 10^{13}\ {\rm G}$) magnetic poles dominate the emission at different accretion rates. On the magnetospheric scale, this configuration is equivalent to an effective dipole field of $\sim 6.6 \times 10^{12}\ {\rm G}$, while allowing the local surface field to exceed $10^{13}\ {\rm G}$.

astro-ph.HE↗

arXiv2026

Investigation on Quasi-periodic Oscillation Phase Lag of RE J1034+396

We conduct an in-depth study of the quasi-periodic oscillation (QPO) properties of RE J1034+396, by constructing QPO phase-folded light curves from 10 XMM-Newton observations during 2020-2021. Our analysis reveals that the QPO in the source exhibits two mutually convertible lag-energy modes: "hard lag" and "soft lag". Despite different lag characteristics, the energy dependency of the root mean square (RMS) amplitude of the QPO under both modes are consistent, suggesting the two types of QPO originate from the same physical mechanism. By performing a spectral analysis, we further find a correlation between time-lag modes and spectral states: the soft lag mode typically corresponds to harder X-ray spectra and higher blackbody temperatures. Through comprehensive comparison of multiple theoretical models, we propose that the relativistic precession model (RPM) of the corona provides a plausible qualitative explanation for the observed complex phenomena, including time-lag mode transitions, and variations of spectral hardness and QPO signal strength.

astro-ph.HE↗

arXiv2026

Off-Policy Actor-Critic with Sigmoid-Bounded Entropy for Real-World Robot Learning

Deploying reinforcement learning in the real world remains challenging due to sample inefficiency, sparse rewards, and noisy visual observations. Prior work leverages demonstrations and human feedback to improve learning efficiency and robustness. However, offline-to-online methods need large datasets and can be unstable, while VLA-assisted RL relies on large-scale pretraining and fine-tuning. As a result, a low-cost real-world RL method with minimal data requirements has yet to emerge. We introduce \textbf{SigEnt-SAC}, an off-policy actor-critic method that learns from scratch using a single expert trajectory. Our key design is a sigmoid-bounded entropy term that prevents negative-entropy-driven optimization toward out-of-distribution actions and reduces Q-function oscillations. We benchmark SigEnt-SAC on D4RL tasks against representative baselines. Experiments show that SigEnt-SAC substantially alleviates Q-function oscillations and reaches a 100\% success rate faster than prior methods. Finally, we validate SigEnt-SAC on four real-world robotic tasks across multiple embodiments, where agents learn from raw images and sparse rewards; results demonstrate that SigEnt-SAC can learn successful policies with only a small number of real-world interactions, suggesting a low-cost and practical pathway for real-world RL deployment.

cs.AI↗

arXiv2026

Vidi2.5: Large Multimodal Models for Video Understanding and Creation

Video has emerged as the primary medium for communication and creativity on the Internet, driving strong demand for scalable, high-quality video production. Vidi models continue to evolve toward next-generation video creation and have achieved state-of-the-art performance in multimodal temporal retrieval (TR). In its second release, Vidi2 advances video understanding with fine-grained spatio-temporal grounding (STG) and extends its capability to video question answering (Video QA), enabling comprehensive multimodal reasoning. Given a text query, Vidi2 can identify not only the corresponding timestamps but also the bounding boxes of target objects within the output time ranges. To enable comprehensive evaluation of STG, we introduce a new benchmark, VUE-STG, which offers critical improvements over existing STG datasets. In addition, we upgrade the previous VUE-TR benchmark to VUE-TR-V2, achieving a more balanced duration and query distribution. Remarkably, the Vidi2 model substantially outperforms leading proprietary systems, such as Gemini 3 Pro Preview and GPT-5, on both VUE-TR-V2 and VUE-STG, while achieving competitive results with popular open-source models with similar scale on video QA benchmarks. The latest Vidi2.5 offers significantly stronger STG capability and slightly better TR and Video QA performance over Vidi2. This update also introduces a Vidi2.5-Think model to handle plot understanding with complex plot reasoning. To comprehensively evaluate the performance of plot understanding, we propose VUE-PLOT benchmark with two tracks, Character and Reasoning. Notably, Vidi2.5-Think outperforms Gemini 3 Pro Preview on fine-grained character understanding with comparable performance on complex plot reasoning. Furthermore, we demonstrate the effectiveness of Vidi2.5 on a challenging real-world application, video editing planning.

cs.CV↗

arXiv2026

Oxygen atom density and kinetics in intermediate-pressure radiofrequency capacitively-coupled plasmas in pure O2

We have studied radiofrequency capacitively coupled plasmas in pure O2 using single mode laser cavity ringdown spectroscopy of oxygen atoms at 630 nm. The absolute atom densities and translational temperatures were determined over a range of pressures and RF power . At pressures of 267 Pa and above, the O atom mole fraction increases with RF power and decreases with pressure, reaching a maximum of 15 percent. However, at 133 and 67 Pa it passes through a distinct maximum with power before decreasing significantly. The atom recombination processes are probed by time resolved measurements in the afterglow of pulse modulated plasmas. At 133 and 67 Pa the atom loss is dominated by surface recombination, and we see clear evidence that this rate is increased by energetic ion bombardment, in agreement with a study from Bill Graham group. This effect partially explains the observed decrease in dissociation at high RF power. The time-resolved results also allow the O negative ion density to be determined and indicate the creation of ozone in the afterglow. At 133 Pa, the trends with RF power of the O2 dissociation, O negative ion density and gas temperature suggest a transition at high power to a plasma mode with fewer high energy electrons. At higher pressures gas phase recombination mechanisms become dominant, however gas convection driven by gas cooling in the afterglow makes it complex to analyse the time-resolved data.

physics.plasm-ph↗

arXiv2026

X-ray and radio observations of the AMXP MAXI J1957+032 covering the 2022-2025 outbursts

We presented a comprehensive multi-epoch timing and multiwavelength analysis of the accreting millisecond X-ray pulsar MAXI J1957+032, covering two major outbursts in 2022 and 2025. By reanalyzing the 2022 outburst data from the Neutron Star Interior Composition Explorer (NICER), we found the spin frequency and orbital parameters from the observations in 0.3-5 keV. For the 2025 outburst, we reported the detection of pulsations with the Einstein Probe (EP). Based on the $\sim$3-year baseline between these two outbursts, we measured a significant long-term spin-down rate of $\dotν= (-5.73 \pm 0.28) \times 10^{-14}~{\rm Hz~s^{-1}}$. Assuming that the quiescent spin-down is driven by magnetic dipole radiation, we inferred a spin-down luminosity of $L \approx 1.1 \times 10^{36}~{\rm erg~s^{-1}}$ and a surface dipolar magnetic field of $B \approx (7.3 - 10.4) \times 10^8$ G. Furthermore, we conducted a deep radio pulsation search with the Five-hundred-meter Aperture Spherical radio Telescope (FAST) during the X-ray quiescent state in 2024, resulting in a non-detection with a 7$σ$ flux density upper limit of 12.3 $μ$Jy. This corresponds to a radio efficiency upper limit of $ξ< 2.8 \times 10^{-10}$, which is significantly lower than that of typical millisecond pulsars with a similar spin-down power. This profound radio pulsation faintness can be explained by two primary scenarios: either a geometric effect, wherein the pulsar's radio beam is directed away from our line of sight, or a physical suppression of the emission mechanism, potentially caused by a persistent low-level accretion flow during the X-ray quiescent state.

astro-ph.HE↗