arXiv subjects

Weiyang Li

Publications and source records attributed to Weiyang Li.

8 recordsLinked to original sources

arXiv2026

Interior gradient estimates for mean curvature equations on Riemannian manifolds

In this paper, we study interior gradient estimates for solutions of \[ \operatorname{div}_{M}\left( \frac{\nabla u}{\sqrt{1+|\nabla u|_g^2}} \right) =f(x,\nabla u) \] on a complete Riemannian manifold of dimension $n\ge2$. Under a Ricci curvature lower bound and a Hessian comparison for the distance function, we derive polynomial and exponential gradient bounds under structural conditions on $f$ and its horizontal and vertical derivatives. The estimates quantify the dependence on the radius, the oscillation of the solution, and the geometric and analytic parameters. In the Euclidean case, we also improve the dependence on the radius in the gradient estimate of Xu~\cite[Theorem~1.1(d)]{Xu2026}. As applications, we obtain Liouville-type results for entire solutions under suitable curvature and growth assumptions.

math.AP↗

arXiv2026

Singular Yamabe metrics on complements of closed sets

Let $n\geq3$ and let $Λ\subset \mathbb S^n$ be closed, contain at least two points, and satisfy $\dim_{\mathrm H}Λ<(n-2)/2$. We prove that there exists a complete metric of scalar curvature $n(n-1)$ on $\mathbb S ^n\setminusΛ$, conformal to the round metric.

math.DG↗

arXiv2026

Dyna3: VLM-Guided Training-Free 4D Reconstruction via Depth Foundation Models

Recent depth foundation models like Depth Anything 3 (DA3) achieve remarkable multi-view depth estimation but assume static 3D scenes, limiting their applicability to real-world dynamic environments. Existing training-free 4D methods like Easi3R and VGGT4D rely on correspondence-trained backbones whose attention encodes cross-frame matching, a property absent in depth-only models like DA3. We present Dyna3, a training-free framework that extends DA3 for 4D dynamic scene reconstruction without any fine-tuning. Our key insight is that DA3's cross-view features, though trained only for depth consistency, implicitly encode motion-discriminative signals when combined with best-match feature search across frames. Its static surfaces find consistent matches globally, while dynamic objects cannot. We further adopt vision-language models (VLM) to automatically generate scene-specific semantic prompts for SAM 3, enabling precise instance-level segmentation that distinguishes which objects move from what objects exist. For reconstruction, we decouple the scene into a cross-frame aligned static background and per-frame dynamic point clouds. Experiments on four datasets demonstrate that Dyna3 surpasses correspondence-trained methods with +5.5pp J-Mean over state-of-the-art VGGT4D on dynamic object segmentation, while achieving up to 13x faster pose estimation and 3x faster 4D reconstruction with 4 to 8x lower memory. Dyna3 could therefore enable much denser temporal sampling that prior methods cannot support.

cs.CV↗

arXiv2026

Model Forensics in AI-Native Wireless Networks: Taxonomy, Applications, and Case Study

As artificial intelligence (AI) is increasingly embedded in wireless networks, models are becoming core components that influence signal processing, resource scheduling and network control. However, model anomalies, tampering and malicious functions also introduce new security risks. In this article, we focus on model forensics in AI-native wireless networks. Specifically, we first discuss key problems including model authenticity verification, malicious function identification and accountability tracing, and summarize the main categories of model forensics. We then explain the role of model forensics in AI-native wireless networks and review representative application scenarios. In the case study, we use RF fingerprinting as an example and present two concrete workflows based on watermark authentication and backdoor detection, illustrating how provenance authentication and malicious behavior identification can be implemented in practice. The results show that model forensics can provide important support for anomaly assessment, provenance tracing and trustworthy operation in AI-native wireless networks. Finally, we outline several promising directions for future research in this emerging area.

cs.CR↗

arXiv2026

OccFace: Unified Occlusion-Aware Facial Landmark Detection with Per-Point Visibility

Accurate facial landmark detection under occlusion remains challenging, especially for human-like faces with large appearance variation and rotation-driven self-occlusion. Existing detectors typically localize landmarks while handling occlusion implicitly, without predicting per-point visibility that downstream applications can benefits. We present OccFace, an occlusion-aware framework for universal human-like faces, including humans, stylized characters, and other non-human designs. OccFace adopts a unified dense 100-point layout and a heatmap-based backbone, and adds an occlusion module that jointly predicts landmark coordinates and per-point visibility by combining local evidence with cross-landmark context. Visibility supervision mixes manual labels with landmark-aware masking that derives pseudo visibility from mask-heatmap overlap. We also create an occlusion-aware evaluation suite reporting NME on visible vs. occluded landmarks and benchmarking visibility with Occ AP, F1@0.5, and ROC-AUC, together with a dataset annotated with 100-point landmarks and per-point visibility. Experiments show improved robustness under external occlusion and large head rotations, especially on occluded regions, while preserving accuracy on visible landmarks.

cs.CV↗

arXiv2025

Advancing Embodied Agent Security: From Safety Benchmarks to Input Moderation

Embodied agents exhibit immense potential across a multitude of domains, making the assurance of their behavioral safety a fundamental prerequisite for their widespread deployment. However, existing research predominantly concentrates on the security of general large language models, lacking specialized methodologies for establishing safety benchmarks and input moderation tailored to embodied agents. To bridge this gap, this paper introduces a novel input moderation framework, meticulously designed to safeguard embodied agents. This framework encompasses the entire pipeline, including taxonomy definition, dataset curation, moderator architecture, model training, and rigorous evaluation. Notably, we introduce EAsafetyBench, a meticulously crafted safety benchmark engineered to facilitate both the training and stringent assessment of moderators specifically designed for embodied agents. Furthermore, we propose Pinpoint, an innovative prompt-decoupled input moderation scheme that harnesses a masked attention mechanism to effectively isolate and mitigate the influence of functional prompts on moderation tasks. Extensive experiments conducted on diverse benchmark datasets and models validate the feasibility and efficacy of the proposed approach. The results demonstrate that our methodologies achieve an impressive average detection accuracy of 94.58%, surpassing the performance of existing state-of-the-art techniques, alongside an exceptional moderation processing time of merely 0.002 seconds per instance.

cs.AI↗

arXiv2023

FeMnNiAlCr High Entropy Alloys with High-Efficiency Surface Oxide Solar Absorbers for Concentrating Solar Power Systems

High entropy alloys (HEAs) have attracted substantial interest in recent years. Thus far, most investigations have focused on their applications as structural materials rather than functional materials. In this paper, we show that FeMnNiAlCr HEAs can potentially be applied as both a structural and functional material for high-efficiency concentrated solar thermal power (CSP) systems working at >700 degrees C. The HEA itself would be used in high-temperature tubing to carry working fluids, while its surface oxide would act as a high-efficiency solar thermal absorber. These HEAs have demonstrated yield strengths 2-3x greater than that of stainless steel at 700 degrees C and a creep lifetime >800 h at 700 degrees C under a typical CSP tubing mechanical load of 35 MPa. Their Mn-rich surface oxides maintain a high optical-to-thermal conversion efficiency of ~87% under 1000x solar concentration for 20 simulated day-night thermal cycles between 750 degrees C and environmental temperature. These HEAs have also sustained immersion in unpurified bromide molten salts for 14 days at 750°C with <2% weight loss, in contrast to 70% weight loss from a 316 stainless steel reference. The simultaneous achievement of promising mechanical, optical, and thermochemical properties in this FeMnNiAlCr system opens the door to new applications of HEAs in solar energy harvesting.

cond-mat.mtrl-sci↗

arXiv2022

3PSDF: Three-Pole Signed Distance Function for Learning Surfaces with Arbitrary Topologies

Recent advances in learning 3D shapes using neural implicit functions have achieved impressive results by breaking the previous barrier of resolution and diversity for varying topologies. However, most of such approaches are limited to closed surfaces as they require the space to be divided into inside and outside. More recent works based on unsigned distance function have been proposed to handle complex geometry containing both the open and closed surfaces. Nonetheless, as their direct outputs are point clouds, robustly obtaining high-quality meshing results from discrete points remains an open question. We present a novel learnable implicit representation, called the three-pole signed distance function (3PSDF), that can represent non-watertight 3D shapes with arbitrary topologies while supporting easy field-to-mesh conversion using the classic Marching Cubes algorithm. The key to our method is the introduction of a new sign, the NULL sign, in addition to the conventional in and out labels. The existence of the null sign could stop the formation of a closed isosurface derived from the bisector of the in/out regions. Further, we propose a dedicated learning framework to effectively learn 3PSDF without worrying about the vanishing gradient due to the null labels. Experimental results show that our approach outperforms the previous state-of-the-art methods in a wide range of benchmarks both quantitatively and qualitatively.

cs.CV↗