arXiv subjects

Xiaoyue Zhang

Publications and source records attributed to Xiaoyue Zhang.

15 recordsLinked to original sources

arXiv2026

STRATA: Self-Learning Through Role-Aligned Tiered Agents for Real-Time Strategy Games

Real-time strategy (RTS) games require agents to coordinate economic development, production and construction, base defense, unit organization, and attack timing over long matches. Existing studies have applied large language models to command decision-making in RTS games, enabling agents to read textual game states and generate high-level plans. However, long inference latency can cause them to miss critical tactical events. The complexity and tactical diversity of full RTS matches also leave existing systems heavily dependent on manually written experience-based prompts, with limited ability to learn continuously from past games. We present STRATA, a role-aligned hierarchical system with cross-game self-learning for Red Alert. STRATA assigns in-game strategic, logistical, and tactical decisions to a Strategic Agent (SA), Logistics Agent (LA), and Tactical Agent (TA), respectively. The SA generates high-level directives based on the global game state and relevant experience cards, while the LA and TA handle logistics and tactical execution. After each match, a Review Agent (RA) derives candidate experience from game traces, validates and revises it using evidence from subsequent matches, and compresses strategic experience supported across multiple games into concise experience cards for SA retrieval. We evaluate STRATA through the formation of experience cards, full-match comparisons before and after learning, and experience learning against AI opponents with different play styles. Under a fixed scenario, using the learned experience cards increases the observed win rate from 30% to 100%. Sequential learning against AI opponents with different play styles also produces distinct long-term strategic experience.

cs.AI↗

arXiv2026

Seeing as Humans Do: Learning from Motion to Segment Anything Without Supervision

The Segment Anything Model (SAM) relies heavily on massive manual annotations, creating a fundamental bottleneck for model scaling. While unsupervised methods attempt to learn object concepts from motion, they typically overfit to moving entities, lacking both multi-granularity understanding and the ability to generalize to static objects. To overcome this, we introduce Motion-Grounded Segment Anything (MoSA), a highly scalable unsupervised framework that learns a transferable objectness prior from unlabeled videos. MoSA operates in three progressive stages: (1) automatically generating multi-granularity motion pseudo-labels from large-scale video data; (2) training a Perceptual Grouping Model (PGM) via contrastive learning to internalize a generalized, appearance-driven concept of objects; and (3) transferring this learned prior into a prompt-guided architecture for segment-anything-style inference on images. Extensive zero-shot evaluations across seven challenging benchmarks (e.g., COCO and ADE20K) demonstrate that MoSA significantly outperforms existing unsupervised methods. Notably, despite using zero manual annotations, MoSA achieves segmentation performance comparable to the fully supervised SAM. Our findings reveal that harnessing large-scale unlabeled motion is a feasible and highly scalable alternative to annotation-driven segment-anything pipelines.

cs.CV↗

arXiv2026

Mechanically Assisted Symmetry Reconstruction for Extraordinary Piezoelectricity

Active symmetry control - a central challenge in materials science, particularly in ferroelectrics - is achieved via mechanically assisted poling (MAP) guided by thermodynamics and phase - field modeling. This approach yields extraordinary piezoelectric coefficients (about 5,000 pC/N at 24 degC; 11,700 pC/N at 58 degC) together with about 65% optical transmittance in a classic relaxor ferroelectric, Pb(Mg1/3Nb2/3)O3-PbTiO3. Mechanical suppression of undesirable phases stabilizes a reconstructed symmetry with highly ordered domains, verified by multiple characterization techniques. The strategy is validated across several distinct ferroelectric systems. To demonstrate its practical utility, we fabricate a transparent dual-modal wearable sensor integrating continuous blood pressure monitoring via piezoelectricity with photoplethysmographic SpO2 detection, enabling high-fidelity physiological tracking. This work establishes mechanically assisted symmetry reconstruction as a pathway to multifunctional optoelectronic materials and compact wearable health technologies.

cond-mat.mtrl-sci↗

arXiv2025

SKYLENAGE Technical Report: Mathematical Reasoning and Contest-Innovation Benchmarks for Multi-Level Math Evaluation

Large language models (LLMs) now perform strongly on many public math suites, yet frontier separation within mathematics increasingly suffers from ceiling effects. We present two complementary benchmarks: SKYLENAGE-ReasoningMATH, a 100-item, structure-aware diagnostic set with per-item metadata on length, numeric density, and symbolic complexity; and SKYLENAGE-MATH, a 150-item contest-style suite spanning four stages from high school to doctoral under a seven-subject taxonomy. We evaluate fifteen contemporary LLM variants under a single setup and analyze subject x model and grade x model performance. On the contest suite, the strongest model reaches 44% while the runner-up reaches 37%; accuracy declines from high school to doctoral, and top systems exhibit a doctoral-to-high-school retention near 79%. On the reasoning set, the best model attains 81% overall, and hardest-slice results reveal clear robustness gaps between leaders and the mid-tier. In summary, we release SKYLENAGE-ReasoningMATH and report aggregate results for SKYLENAGE-MATH; together, SKYLENAGE provides a hard, reasoning-centered and broadly covering math benchmark with calibrated difficulty and rich metadata, serving as a reference benchmark for future evaluations of mathematical reasoning.

cs.CL↗

arXiv2025

Girlhood Feminism as Soft Resistance: Affective Counterpublics and Algorithmic Negotiation on RedNote

This article explores how Chinese female users tactically mobilise platform features and hashtag practices to construct vernacular forms and an exclusive space of feminist resistance under algorithmic and cultural constraints. Focusing on the reappropriation of the hashtag Baby Supplementary Food (BSF), a female-dominated lifestyle app with over 300 million users, we analyse how users create a female-centered counterpublic through self-infantilisation, algorithmic play, and aesthetic withdrawal. Using the Computer-Assisted Learning and Measurement (CALM) framework, we analysed 1580 posts and propose the concept of girlhood feminism: an affective, culturally grounded form of soft resistance that refuses patriarchal life scripts without seeking direct confrontation or visibility. Rather than challenging censorship and misogyny directly, users rework platform affordances and domestic idioms to carve out emotional and symbolic spaces of dissent. Situated within the broader dynamics of East Asia's compressed modernity, this essay challenges liberal feminist paradigms grounded in confrontation and transparency. It advances a regionally grounded framework for understanding how gendered publics are navigated, negotiated, and quietly reimagined in algorithmically governed spaces.

cs.CY↗

arXiv2024

Creation of independently controllable and long lifetime polar skyrmion textures in ferroelectric-metallic heterostructures

Topological textures like vortices, labyrinths and skyrmions formed in ferroic materials have attracted extensive interests during the past decade for their fundamental physics, intriguing topology, and technological prospects. So far, polar skyrmions remain scarce in ferroelectrics as they require a delicate balance between various dipolar interactions. Here, we report that PbTiO3 thin films in a metallic contact undergo a topological phase transition and stabilize a broad family of skyrmion-like textures (e.g., skyrmion bubbles, multiple π-twist target skyrmions, and skyrmion bags) with independent controllability, analogous to those reported in magnetic systems. Weakly-interacted skyrmion arrays with a density over 300 Gb/inch2 are successfully written, erased and read-out by local electrical and mechanical stimuli of a scanning probe. Interestingly, in contrast to the relatively short lifetime <20 hours of the skyrmion bubbles, the multiple π-twist target skyrmions and skyrmion bags show topology-enhanced stability with lifetime over two weeks. Experimental and theoretical analysis implies the heterostructures carry electric Dzyaloshinskii-Moriya interaction mediated by oxygen octahedral tiltings. Our results demonstrate ferroelectric-metallic heterostructures as fertile playground for topological states and emergent phenomena.

cond-mat.mtrl-sci↗

arXiv2024

Online Resource Allocation with Non-Stationary Customers

We propose a novel algorithm for online resource allocation with non-stationary customer arrivals and unknown click-through rates. We assume multiple types of customers arrive in a nonstationary stochastic fashion, with unknown arrival rates in each period, and that customers' click-through rates are unknown and can only be learned online. By leveraging results from the stochastic contextual bandit with knapsack and online matching with adversarial arrivals, we develop an online scheme to allocate the resources to nonstationary customers. We prove that under mild conditions, our scheme achieves a ``best-of-both-world'' result: the scheme has a sublinear regret when the customer arrivals are near-stationary, and enjoys an optimal competitive ratio under general (non-stationary) customer arrival distributions. Finally, we conduct extensive numerical experiments to show our approach generates near-optimal revenues for all different customer scenarios.

cs.LG↗

arXiv2023

Observation of Fluctuation Spin Hall Effect in Antiferromagnet

The spin Hall effect (SHE) can generate a pure spin current by an electric current, which is promisingly used to electrically control magnetization. To reduce power consumption of this control, a giant spin Hall angle (SHA) in the SHE is desired in low-resistivity systems for practical applications. Here, critical spin fluctuation near the antiferromagnetic (AFM) phase-transition is proved as an effective mechanism to create an additional part of SHE, named as fluctuation spin Hall effect (FSHE). This FSHE enhances the SHA due to the AFM spin fluctuation between conduction electrons and local spins. We detect the FSHE with the inverse and direct spin Hall effect (ISHE and DSHE) set-up and their temperature (T) dependences in the Cr/MgO/Fe magnetic tunnel junctions (MTJs). The SHA is significantly enhanced when temperature is approached to the Néel temperature (T_N) and has a peak value of -0.34 at 200 K near T_N. This value is higher than the room-temperature value by 240% and comparable to that of heavy metals Ta and W. Furthermore, the spin Hall resistivity of Cr well fits the modeled T-dependence when T approaches T_N from low temperatures, implying the AFM spin fluctuation nature of strong SHA enhancement. Thus, this study demonstrates the critical spin fluctuation as a prospective way of increasing SHA and enriches the AFM material candidates for spin-orbitronic devices.

cond-mat.mes-hall↗

arXiv2023

The appearance of a merging binary black hole very close to a spinning supermassive black hole

The mass and distance of a binary black hole (BBH) are fundamental parameters to measure in gravitational-wave (GW) astronomy. It is well-known that the measurement is affected by cosmological redshift, and recent works also showed that Doppler and gravitational redshifts could further affect the result if the BBH coalesces close to a supermassive black hole (SMBH). Here we consider the additional lensing effect induced by the nearby SMBH on the measurement. We compute the null geodesics originating within $10$ gravitational radii of a Kerr SMBH to determine the redshift and magnification of the GWs emitted by the BBH. We find a positive correlation between redshift and demagnification, which results in a positive correlation between the mass and distance of the BBH in the detector frame. More importantly, we find a higher probability for the signal to appear redshifted and demagnified to a distant observer, rather than blueshifted and magnified. Based on these results, we show that a binary at a cosmological redshift of $z_{\rm cos}=(10^{-2}-10^{-1})$ and composed of BHs of $(10-20)M_\odot$ could masquerade as a BBH at a redshift of $z_{\rm cos}\sim1$ and containing BHs as large as $(44-110)M_\odot$. In the case of extreme demagnification, we also find that the same BBH could appear to be at $z_{\rm cos}>10$ and contain subsolar-mass BHs. Such an effect, if not accounted for, could bias our understanding of the origin of the BHs detected via GWs.

astro-ph.HE↗

arXiv2022

Cross-domain Detection Transformer based on Spatial-aware and Semantic-aware Token Alignment

Detection transformers like DETR have recently shown promising performance on many object detection tasks, but the generalization ability of those methods is still quite challenging for cross-domain adaptation scenarios. To address the cross-domain issue, a straightforward way is to perform token alignment with adversarial training in transformers. However, its performance is often unsatisfactory as the tokens in detection transformers are quite diverse and represent different spatial and semantic information. In this paper, we propose a new method called Spatial-aware and Semantic-aware Token Alignment (SSTA) for cross-domain detection transformers. In particular, we take advantage of the characteristics of cross-attention as used in detection transformer and propose the spatial-aware token alignment (SpaTA) and the semantic-aware token alignment (SemTA) strategies to guide the token alignment across domains. For spatial-aware token alignment, we can extract the information from the cross-attention map (CAM) to align the distribution of tokens according to their attention to object queries. For semantic-aware token alignment, we inject the category information into the cross-attention map and construct domain embedding to guide the learning of a multi-class discriminator so as to model the category relationship and achieve category-level token alignment during the entire adaptation process. We conduct extensive experiments on several widely-used benchmarks, and the results clearly show the effectiveness of our proposed method over existing state-of-the-art baselines.

cs.CV↗

arXiv2020

Deform-GAN:An Unsupervised Learning Model for Deformable Registration

Deformable registration is one of the most challenging task in the field of medical image analysis, especially for the alignment between different sequences and modalities. In this paper, a non-rigid registration method is proposed for 3D medical images leveraging unsupervised learning. To the best of our knowledge, this is the first attempt to introduce gradient loss into deep-learning-based registration. The proposed gradient loss is robust across sequences and modals for large deformation. Besides, adversarial learning approach is used to transfer multi-modal similarity to mono-modal similarity and improve the precision. Neither ground-truth nor manual labeling is required during training. We evaluated our network on a 3D brain registration task comprehensively. The experiments demonstrate that the proposed method can cope with the data which has non-functional intensity relations, noise and blur. Our approach outperforms other methods especially in accuracy and speed.

cs.CV↗

arXiv2019

Fixed points with finite mean of the smoothing transform in random environments

At each time $n\in\mathbb{N}$, let $\bar{Y}^{(n)}=(y_{1}^{(n)},y_{2}^{(n)},\cdots)$ be a random sequence of non-negative numbers that are ultimately zero in a random environment $ξ=(ξ_{n})_{n\in\mathbb{N}}$ in time, which satisfies for each $n\in\mathbb{N}$ and a.e. $ξ,~E_ξ[\sum_{i\in\mathbb{N}_{+}}y_{i}^{(n)}(ξ)]=1.$ The existence and uniqueness of the non-negative fixed points of the associated smoothing transform in random environments is considered. These fixed points are solutions of the distributional equation for $a.e.~ξ,~Z(ξ)\overset{d}{=}\sum_{i\in\mathbb{N}_{+}}y_{i}^{(0)}(ξ)Z_{i}(Tξ),$ where when given the environment $ξ$, $Z_{i}(Tξ)~(i\in\mathbb{N}_{+})$ are $i.i.d.$ non-negative random variables, and distributed the same as $Z(ξ)$. As an application, the martingale convergence of the branching random walk in random environments is given as well. The classical results by Biggins (1977) has been extended to the random environment situation.

math.PR↗

arXiv2018

Asymptotic behaviour of heavy-tailed branching processes in random environments

Consider a heavy-tailed branching process (denoted by $Z_{n}$) in random environments, under the condition which infers that $\mathbb{E}\log m(ξ_{0})=\infty$. We show that (1) there exists no proper $c_{n}$ such that $\{Z_{n}/c_{n}\}$ has a proper, non-degenerate limit, (2) normalized by a sequence of functions, a proper limit can be obtained, i.e., $y_{n}\left(\barξ,Z_{n}(\barξ)\right)$ converges almost surely to a random variable $Y(\barξ)$, where $Y\in(0,1)~η$-a.s., (3) finally, we give a necessary and sufficient conditions for the almost sure convergence of $\left\{\frac{U(\barξ,Z_{n}(\barξ))}{c_n(\barξ)}\right\}$, where $U(\barξ)$ is a slowly varying function that may depends on $\barξ$.

math.PR↗

arXiv2018

Limit theorems for the minimal position of a branching random walk in random environment

We consider a branching system of random walk in random environment (in location) in $\mathbb{N}$. We will give the exact limit value of $\frac{M_{n}}{n}$, where $M_{n}$ denotes the minimal position of branching random walk at time $n$. A key step in the proof is to transfer our branching random walks in random environment (in location) to branching random walks in random environment (in time), by use of Bramson's "branching processes within a branching process" .

math.PR↗

arXiv2018

Slowly rotating Bose Einstein Condensate galactic dark matter halos, and their rotation curves

If dark matter is composed of massive bosons, a Bose-Einstein Condensation process must have occurred during the cosmological evolution. Therefore galactic dark matter may be in a form of a condensate, characterized by a strong self-interaction. We consider the effects of rotation on the Bose-Einstein Condensate dark matter halos, and we investigate how rotation might influence their astrophysical properties. In order to describe the condensate we use the Gross-Pitaevskii equation, and the Thomas-Fermi approximation, which predicts a polytropic equation of state with polytropic index $n=1$. By assuming a rigid body rotation for the halo, with the use of the hydrodynamic representation of the Gross-Pitaevskii equation we obtain the basic equation describing the density distribution of the rotating condensate. We obtain the general solutions for the condensed dark matter density, and we derive the general representations for the mass distribution, boundary (radius), potential energy, velocity dispersion, tangential velocity and for the logarithmic density and velocity slopes, respectively. Explicit expressions for the radius, mass, and tangential velocity are obtained in the first order of approximation, under the assumption of slow rotation. In order to compare our results with the observations we fit the theoretical expressions of the tangential velocity of massive test particles moving in rotating Bose-Einstein Condensate dark halos with the data of 12 dwarf galaxies, and the Milky Way, respectively.

gr-qc↗