arXiv subjects

Song Zhou

Publications and source records attributed to Song Zhou.

15 recordsLinked to original sources

arXiv2026

Colossal reversible conductivity switching by room-temperature oxygen-vacancy ordering in Aurivillius oxide films

Oxygen vacancies are central to the functionality of oxides, yet they typically exist as randomly distributed point defects, limiting the ability to precisely manipulate their collective behavior. Here, we report the room-temperature formation of a long-range-ordered oxygen-vacancy superstructure in single-crystalline Aurivillius-phase Bi2WO6 thin films via a mild nitrogen-plasma treatment. This structural transformation unlocks a colossal, reversible modulation of electrical conductivity by more than nine orders of magnitude, accompanied by a striking optical transition from transparent to black. Atomic-resolution imaging and spectroscopy reveal that the vacancies selectively order within the perovskite-like tungsten oxide layers, forming a coherent defect lattice that is absent in the pristine film. Oxygen-plasma treatment removes the vacancy superstructure and restores the initial state, whereas subsequent nitrogen-plasma treatment reconstructs it, enabling repeatable room-temperature switching between distinct structural, electronic and optical states. The phenomenon is also observed in another Aurivillius member, Bi2MoO6, suggesting its generality across the Aurivillius family. These findings establish a new paradigm for atomic-scale defect engineering - using gentle plasma chemistry to construct ordered defect lattices, opening avenues for reversible property modulation in complex oxides.

cond-mat.mtrl-sci↗

arXiv2026

Action-grounded tissue affordance enables anticipatory auto-framing that lowers surgeon cognitive workload during laparoscopic surgery

In laparoscopy, surgeon gaze tracks where the instruments will act; easing this demand through visual attention modeling requires dense labels of those interaction loci. These encode tacit knowledge: experts converge on consensus loci yet struggle to state the rules. Here we show that such labels can be recovered from completed actions in surgical videos, in which recorded instrument trajectories are converted into dense, continuous supervision. DiffeoAfford grounds tissue affordance by attaching instrument tips to the tissue and transporting them through deformation using diffeomorphism-constrained tracking, matching context-informed annotators' accuracy. Trained on these labels and never on gaze, a real-time model aligns with surgeon gaze more closely in space and time than does camera-assistant gaze. The framework also transfers across procedures: on hysterectomy videos, a separately trained predictor reaches 95.16% directional consistency with subsequent camera motion. In 12 paired cholecystectomies (24 procedures), the auto-framing application AffordView, which proactively centers predicted targets in view, lowered surgeon cognitive workload on converging subjective, physiological, and behavioral measures, including a reduced number of verbal instructions to the camera assistant. Deriving supervision from action rather than manual annotation offers a scalable route to anticipatory assistance.

cs.CV↗

arXiv2026

LoopFM: Learning frOm HistOrical RePresentations of Foundation Model for Recommendation

Knowledge distillation (KD) transfers a single scalar prediction from a large foundation model (FM) to compact vertical models (VMs), suffering from diminishing transfer ratio -- the fraction of FM improvement captured by the VM -- as a single scalar cannot convey the rich intermediate knowledge that larger FMs learn. To address this bottleneck, we propose LoopFM (Learning frOm HistOrical RePresentations of FM), a framework that opens a high-bandwidth transfer channel by structuring FM intermediate embeddings as input features (e.g., user history sequence) for downstream VMs, without requiring real-time FM inference at serving and architectural coupling between FM and VM. We provide a theoretical framework for LoopFM with a gain decomposition and transfer-ratio analysis. On three public benchmarks, LoopFM demonstrates strong AUC improvements (e.g., 6%+ on TaobaoAd) and complementary knowledge transfer capability with KD. On industrial-scale systems (billions of examples, trillion-parameter FMs), LoopFM approximately doubles the knowledge transfer ratio on top of KD, delivering a +0.5% conversion improvement in the first half after its initial launch, and +1.03% and +1.22% conversion improvement from two individual launches in the subsequent half.

cs.LG↗

arXiv2025

External Large Foundation Model: How to Efficiently Serve Trillions of Parameters for Online Ads Recommendation

Ads recommendation is a prominent service of online advertising systems and has been actively studied. Recent studies indicate that scaling-up and advanced design of the recommendation model can bring significant performance improvement. However, with a larger model scale, such prior studies have a significantly increasing gap from industry as they often neglect two fundamental challenges in industrial-scale applications. First, training and inference budgets are restricted for the model to be served, exceeding which may incur latency and impair user experience. Second, large-volume data arrive in a streaming mode with data distributions dynamically shifting, as new users/ads join and existing users/ads leave the system. We propose the External Large Foundation Model (ExFM) framework to address the overlooked challenges. Specifically, we develop external distillation and a data augmentation system (DAS) to control the computational cost of training/inference while maintaining high performance. We design the teacher in a way like a foundation model (FM) that can serve multiple students as vertical models (VMs) to amortize its building cost. We propose Auxiliary Head and Student Adapter to mitigate the data distribution gap between FM and VMs caused by the streaming data issue. Comprehensive experiments on internal industrial-scale applications and public datasets demonstrate significant performance gain by ExFM.

cs.IR↗

arXiv2025

Healing of the edge magnetic island in the island divertor configuration on J-TEXT

The phenomena of island healing and configuration transition induced by high-power electron cyclotron resonance heating (ECRH) have been investigated in the island divertor configuration on the J-TEXT tokamak. Experimental results reveal that the size of the edge open magnetic island with mode number m/n = 3/1 decreases substantially under specific ECRH conditions. This process, referred to as island healing, occurs when ECRH with a power of 500~600 kW is deposited in the plasma core or when 250 kW of ECRH is deposited at r = 0.5 a, where a is the minor radius. The reduction of the island width makes the island divertor ineffective and transition into the limiter configuration. A model incorporating the influence of ECRH on the scrape-off layer (SOL) thermoelectric current is proposed to explain the observed changes in the edge magnetic topology of the island divertor configuration. These findings suggest that ECRH should be deposited at the plasma core with carefully controlled power to ensure the stable and compatible operation of ECRH and the island divertor configuration in tokamaks. The results can provide insights into achieving robust operation of an island divertor in tokamaks.

physics.plasm-ph↗

arXiv2024

Leveraging Surgical Activity Grammar for Primary Intention Prediction in Laparoscopy Procedures

Surgical procedures are inherently complex and dynamic, with intricate dependencies and various execution paths. Accurate identification of the intentions behind critical actions, referred to as Primary Intentions (PIs), is crucial to understanding and planning the procedure. This paper presents a novel framework that advances PI recognition in instructional videos by combining top-down grammatical structure with bottom-up visual cues. The grammatical structure is based on a rich corpus of surgical procedures, offering a hierarchical perspective on surgical activities. A grammar parser, utilizing the surgical activity grammar, processes visual data obtained from laparoscopic images through surgical action detectors, ensuring a more precise interpretation of the visual information. Experimental results on the benchmark dataset demonstrate that our method outperforms existing surgical activity detectors that rely solely on visual features. Our research provides a promising foundation for developing advanced robotic surgical systems with enhanced planning and automation capabilities.

cs.RO↗

arXiv2023

Photonic Jet with Tunable Focus Based on Water Droplets Freezing from the Outside In

Water droplets are a perspective highly abundant phase-change material to realize tunable optical lenses. We demonstrated for the first time that freezing mesoscale water droplet could be use as tunable optical lens, such that freezing becomes an asset despite both the low absolute values of the refractive indices of the shell and core materials and their optical contrast. It was shown that the dielectric shell of mesoscale water droplet in the form of solid ice allows controlling both the maximum field intensity and the focus position of the formed photonic nanojet. The formation of ice with air bubbles during the freezing of a water droplet is appropriate for a dynamic increase in the range of change of the focal position compared to solid ice. The proposed concept of a tunnelable spherical lens based on a freezing water drop can be used for microscopy, optical traping in "green" mesotronics.

physics.optics↗

arXiv2022

Cascades of Fano Resonances in Scattering by a Mesoscale Spherical Particle in the Superresonance Mode

Broadband light illumination of a mesoscale dielectric sphere makes it possible to reveal new effects that associated with super resonance mode. These include the possibility of generating high-order Fano resonance cascades. The quality factor has the order Q=10^7. Super-resonance-enabled subdiffraction fields localization by mesoscale dielectric sphere under broadband light illuminations (e.g., at wavebands of 400 - 700 nm for refractive index of sphere n=1.5-1.9 and 1500 - 1600 nm for n=3.47) have been investigated. The conditions for the quasi-periodicity of superresonance peaks are established. The dependence of the amplitude modulation of Fano cascades on the refractive index of the sphere is shown. These results are important in deep understanding of physics of the super-resolution mechanism related to superresonance mode and will find great potential applications in many other area.

physics.optics↗

arXiv2022

Super-resonance effect for high-index sphere immersed in water

Recently, we showed that dielectric mesoscale spheres support super-resonance effect, i.e. high-order Mie resonance modes with giant field enhancement. The presence of the surrounding medium leads to a significant influence in the intensity of the electric and magnetic fields in the particle. In this paper, we show that this effect can be used for highly precise control of the effective refractive index of a medium, such as water. We show that a change in the water temperature by dT=0.0106 C (or the effective refractive index of the medium by 2e-6) leads to a twofold drop in electric field intensity. All the presented results are strictly within the framework of the classical Mie theory without any modifications. A detailed study of the ranges of values of the size parameters of a spherical particle of the order of 10, which had previously been neglected, made it possible to reveal a new, unusual physics of the phenomenon.

physics.optics↗

arXiv2022

The super resonance effect paves the way for a new type of refractive index sensor concept based on a mesoscale dielectric sphere

Recently, we showed that dielectric mesoscale spheres could support so-called super-resonance effect, i.e. high-order Mie resonance modes with giant field localization and enhancement. Due to the presence of the surrounding medium leads to a significant influence in the intensity of the field in the particle, based on Mie theory we show for the first time that this effect may be use to design refractive index sensor of medium. Using the example of air as an environment, we have shown that the sensitivity of the proposed sensor concept can reach from 10-6 to 10-8, depending on the accuracy of the sphere size parameter, which is no worse than the accuracy of modern interference methods.

physics.optics↗

arXiv2022

Influence of the Environment on the Effect of Super Resonance in Mesoscale Dielectric Spheres

Dielectric mesoscale spheres have aroused strong interest because of their potential to localize light at deep subwavelength volume and to yield extremal internal magnetic and/or electric field enhancements. Recently, we showed that such particle could support high-order Mie resonance modes with giant field localization and enhancement. Optimizing the internal fields appears as a key challenge for enhancing wave matter interactions in dielectric mesoscale particles. However, a dielectric particle is always located in some medium, and not in a vacuum. Moreover, the question is how much the environment medium affects the internal field intensities enhancement in the super-resonance effect. Based on Mie theory we show for the first time that the presence of the environment leads to a significant decrease in the intensity of the field in the particle. Thus, the study of the effect of super-resonance becomes meaningless without taking into account the environment. However, a greater enhancement of the internal field is found for the blue-shifted Mie size parameter of the sphere when the particle, for example, is in air rather than in vacuum.

physics.optics↗

arXiv2022

Magnetic hot-spots generation at optical frequencies in all-dielectric mesoscale Janus particles

At optical frequencies due to the small value of the magnetic permeability of natural materials, the magnetic effects are week. To this end, the natural dielectric materials are unemployable for practical magnetic applications in optics. We have shown that it is possible to induce the intense magnetic hot spots in a Janus dielectric mesoscale particle. The basic idea of the Janus particle based on a combination of the effects of a photonic jet, whispering gallery waves and the concept of solid immersion. Simulations show that H^2/E^2 contrast maybe more 10 and maximal magnetic field intensity enhancement is more than 1000 for a wavelength-scaled particle with refractive index less than 2.

physics.optics↗

arXiv2022

Subwavelength field localization based on dielectric mesoscale particle with single and blind nanohole array

Some new unusual physical phenomena and effects associated with dielectric mesoscale particles with Mie size parameter near 10 were studied and have been discovered during the last decade. In this paper, we propose nanoholes structured wavelength-scaled dielectric cubic particle with refractive index near two, where the array of nanoholes can act as a plurality of near-field probes to simultaneously illuminate the sample surface and it has the potential of surpassing the performance of most existing nearfield imaging approaches. We also offer the concept of the single nano-structuring of a dielectric cylinder or sphere made from conventional optical materials. The choice of the diameter of the nanohole in the particle "transfers" it into the resonance mode, when the characteristics of the field localized in the shadow part of the particle are determined not by the wavelength, but by the size of the nanohole. Thus, the diameter of the focused spot at the exit from the particle can be much smaller than the solid immersion diffraction limit.

physics.optics↗

arXiv2021

A Different Perspective On The Stochastic Convex Feasibility Problem

We analyze a simple randomized subgradient method for approximating solutions to stochastic systems of convex functional constraints, the only input to the algorithm being the size of minibatches. By introducing a new notion of what is meant for a point to approximately solve the constraints, determining bounds on the expected number of iterations reduces to determining a hitting time for a compound Bernoulli process, elementary probability. Besides bounding the expected number of iterations quite generally, we easily establish concentration inequalities on the number of iterations, and more interesting, we establish much-improved bounds when a notion akin to H\"{o}lderian growth is satisfied, for all degrees of growth, not just the linear growth of piecewise-linear convex functions or the quadratic growth of strongly convex functions. Finally, we establish the analogous results under a slight modification to the algorithm which results in the user knowing with high confidence an iterate is in hand that approximately solves the system. Perhaps surprisingly, the iteration bounds here are deterministic -- all of the probability gets wrapped into the confidence level (albeit at the expense of potentially large minibatches).

math.OC↗

arXiv2018

Limited Memory Kelley's Method Converges for Composite Convex and Submodular Objectives

The original simplicial method (OSM), a variant of the classic Kelley's cutting plane method, has been shown to converge to the minimizer of a composite convex and submodular objective, though no rate of convergence for this method was known. Moreover, OSM is required to solve subproblems in each iteration whose size grows linearly in the number of iterations. We propose a limited memory version of Kelley's method (L-KM) and os OSM that requires limited memory (at most n + 1 constraints for an n-dimensional problem) independent of the iteration. We prove convergence for L-KM when the convex part of the objective (g) is strongly convex and show it converges linearly when g is also smooth. Our analysis relies on duality between minimization of the composite objective and minimization of a convex function over the corresponding submodular base polytope. We introduce a limited memory version, L-FCFW, of the Fully-Corrective Frank-Wolfe (FCFW) method with approximate correction, to solve the dual problem. We show that L-FCFW and L-KM are dual algorithms that produce the same sequence of iterates; hence both converge linearly (when g is smooth and strongly convex) and with limited memory. We propose L-KM to minimize composite convex and submodular objectives; however, our results on L-FCFW hold for general polytopes and may be of independent interest.

math.OC↗