Searcharxiv⌕ Search

arXiv subjects

Qingsong Wang

Publications and source records attributed to Qingsong Wang.

At least 19 recordsLinked to original sources

Dual-Core Tensor Wheel Decomposition with Structure-Cached Acceleration for Tensor Recovery

Tensor wheel decomposition captures complex correlations in high-dimensional data through interactions between outer-ring factors and a high-order central core. However, updating these components requires the explicit construction and manipulation of large-scale matrices, which limits computational efficiency and incurs particularly high computational and memory costs for the central-core update. To address these limitations, we propose dual-core tensor wheel decomposition (DCTW), which factorizes the central core into two lower-order cores linked by a shared latent bond. Each core is coupled to a different group of outer-ring factors. We further propose a structure-cached acceleration strategy that avoids explicit large-scale matrix multi- plications and caches intermediate tensors generated during tensor network contractions for reuse across block updates. This reuse reduces redundant computation and avoids the explicit construction of large matrices. Complexity analysis and experiments show that the strategy based on structural caching accelerates optimization compared with explicit and implicit up- date schemes while limiting the increase in peak memory as the ranks grow. Experiments on color video completion and robust denoising show that DCTW matches or outperforms representative tensor network methods in reconstruction quality, computational efficiency, and scalability.

cs.LG↗

Critical Diameters and Vietoris-Rips Filtrations

We relate critical diameters of finite configurations to the topology of Vietoris-Rips complexes. For a compact metric space and $0<r<s$, we prove that the canonical inclusion from scale $r$ to scale $s$ is a homotopy equivalence whenever $[r,s)$ contains no diameter of a finite labelled configuration at which the diameter function has zero weak slope. On Riemannian manifolds, weak-slope stationarity implies Clarke criticality. At positive diameter, when the distances realizing the diameter are smooth, both are equivalent to first-order stationarity: the absence of a direction decreasing all these distances to first order. For closed connected smooth manifolds, the Clarke critical diameter spectrum has Hausdorff dimension zero, even at the cut locus. In the real-analytic case, it is finite at each fixed number of labels and countable over all label numbers. On the unit round sphere $S^m$, $m\geq1$, positive nonantipodal weak-slope stationary configurations are characterized by nonzero nonnegative equilibrium stresses. The least positive critical diameter is $\arccos(-1/(m+1))$, and the first accumulation point is $\arccos(-1/m)$. We lift every such stress through a spherical stack construction adapted from Lovász, producing stationary diameters that approach the original value from below in the next dimension as the number of layers increases. We also give example in general metric space showing that weak-slope stationary diameters need not correspond to changes in the homotopy type of Vietoris-Rips complexes.

math.AT↗

Small deviations for integrators in terms of Itô--Wiener expansions

In this paper, we investigate small deviations for a class of Gaussian processes, known as integrators. In general, such processes are non-Markovian and may even fail to be semimartingales. First, we investigate small deviations for the multi-dimensional Wiener process, focusing on the asymptotic behavior of the Fourier--Wiener transform of the associated hitting event. We analyze the Fourier--Wiener transform and the separate summands in its Itô--Wiener expansion. Then, we establish the asymptotic limits of the associated multiple integrals with heat kernels. Finally, we obtain the small deviation asymptotic result for the integrator $η$ associated with the continuous linear operator.

math.PR↗

Gromov-Hausdorff Distances Between Spheres: Asymptotics and Vietoris-Rips Coindex Bounds

For $1\le m<n$, let $c_{m,n}$ be the infimum of the scales $r$ at which the Vietoris-Rips simplicial complex of the unit round sphere $\mathbb{S}^m$, equipped with its geodesic metric, admits a continuous odd map from $\mathbb{S}^n$. Previous work established the lower bound $\frac{1}{2}c_{m,n}\le d_{\mathrm{GH}}(\mathbb{S}^m,\mathbb{S}^n)$ through a quantitative Borsuk-Ulam theorem and raised the question of whether it is always sharp. We show that this bound can underestimate the distance by a factor arbitrarily close to two, the largest possible factor. Combining projective-code upper bounds for $c_{m,n}$ with lower bounds for $d_{\mathrm{GH}}$ obtained from obstructions to homomorphisms between Borsuk graphs, we obtain explicit finite-dimensional bounds. Along sequences of pairs $1\le m<n$, we have $c_{m,n}\to\fracπ{2}$ whenever $\log n/m\to0$, while $d_{\mathrm{GH}}(\mathbb{S}^m,\mathbb{S}^n)\to\fracπ{2}$ whenever $m/n\to0$. When both conditions hold (for example when $n=m^2$) the ratio of the distance to $\frac{1}{2}c_{m,n}$ tends to two. For each fixed $m\ge3$, the inequality is strict for all sufficiently large $n$.

math.MG↗

A Multi-Step Inertial ADMM Algorithm for Federated Learning

Federated learning has recently emerged as an important paradigm for large-scale distributed model training, which can be naturally formulated as a class of distributed learning problems. The alternating direction method of multipliers (ADMM) is a widely used approach for solving such problems. However, its practical performance is often hindered by significant communication overhead and system heterogeneity. In this paper, we propose a multi-step inertial ADMM algorithm, called MiFedADMM, for federated learning. The proposed method leverages multiple historical iterates on the server side to construct a higher-order extrapolation after each aggregation to accelerate the convergence behavior. Under mild assumptions, we establish convergence results for the single-step variant of the proposed method and derive an O(1/K) convergence rate. Numerical experiments on both synthetic and real-world datasets demonstrate the effectiveness and improved performance of the proposed method compared with existing approaches.

cs.LG↗

Composite and Stochastic Smoothed Decentralized Bregman Proximal-Gradient Methods

Smoothed decentralized Bregman proximal-gradient (SDBPG) methods combine relative-smoothness geometry with edge-dual consensus. We extend SDBPG to node-specific convex nonsmooth regularizers and unbiased stochastic local gradients while retaining its Bregman primal-dual smoothing mechanism. The analysis isolates temporal subgradient motion and aggregation mismatch as two defect sequences. For trajectories in a forward-invariant compact subset of the kernel domain and admissible parameters, finite accumulated defects yield an $O(T^{-1})$ bound on the squared best-iterate proximal residual over $T$ iterations. The stochastic bound separates accumulated oracle variance from expected defects and attains the same order when the variance bounds and expected defects are summable. Common active-structure identification, together with local subgradient conditions, gives sufficient finite-defect conditions for $\ell_1$ and group $\ell_{2,1}$ regularization. Experiments with common regularizers and heterogeneous sensing operators relate defect decay to observed support stabilization and compare reconstruction and communication with decentralized proximal and gradient-tracking methods. A separate phase-retrieval study evaluates the effect of a quartic Bregman kernel. Code and experiment scripts are available at https://github.com/shengze-xu/composite-stochastic-sdbpg .

math.OC↗

Emotion Understanding in Streaming Video with Trajectory-Aware Reliability

Video emotion understanding is commonly studied as an offline classification problem, where the complete video segment is available before prediction. Real-time interaction, however, requires emotion decisions from incomplete and evolving evidence. This paper studies streaming video emotion understanding as a reliability-aware decision process over evolving emotion beliefs. In this setting, a single confident prefix prediction can still be unreliable when the underlying belief trajectory is unstable or repeatedly switches across emotion classes. We propose TRACE, a trajectory-aware reliability framework that forms low-latency emotion beliefs from streaming audio prefixes, estimates reliability from confidence, entropy, stability, and class-switching patterns, and selectively invokes contextual belief reinterpretation with visual, textual, and neighboring-utterance evidence. TRACE keeps stable cases in the low-latency online pathway while allocating stronger multimodal reasoning to uncertain cases that remain ambiguous. Experiments on StreamMER, MELD, and MER2024 show that TRACE improves the accuracy-cost trade-off, retaining most full-context gains while reducing unnecessary contextual reasoning.

cs.MM↗

Balanced Bismut torsion-parallel manifold with constant holomorphic sectional curvature

A long-lasting conjecture in non-Kähler geometry says that any compact Hermitian manifold with constant holomorphic sectional curvature must be either Kähler or Chern flat. The conjecture is known to be true in dimension 2 but remains open in dimensions 3 or higher, except in several special cases. In this article we consider compact Hermitian manifolds that are Bismut torsion-parallel (BTP for brevity), which means that the Bismut connection has parallel torsion. For BTP manifolds that are not balanced, the conjecture was proved by Chen and Zheng. They also confirmed the conjecture for balanced BTP manifolds in dimension 3, utilizing a classification result for such threefolds by Zhao and Zheng. In a recent work, we confirmed the conjecture for balanced BTP manifolds in dimension 4 through a detailed case-by-case analysis. In this article, we establish the conjecture for balanced BTP manifolds in all dimensions. We also disprove the full-rank $B$ conjecture by constructing a compact balanced BTP fivefold with full-rank $B$ that is neither Kähler nor Chern flat; its Chern holomorphic sectional curvature is nonconstant.

math.DG↗

On Compact Hermitian Surfaces With Positive Strominger-Bismut Sectional Curvature

In an earlier work, Yau and Zheng proposed a Hermitian analogue of a weak form of the Frankel conjecture, which states that any compact Hermitian manifold with positive Strominger-Bismut sectional curvature must be biholomorphic to the complex projective space. In this article, we confirm the conjecture in complex dimension two. This is achieved by a vanishing theorem for anti-self-dual harmonic 2-forms on such surfaces, utilizing a curvature decomposition formula by Ferreira and a refined Kato inequality.

math.DG↗

Balanced Bismut torsion-parallel fourfolds with constant holomorphic sectional curvature

An old conjecture in non-Kähler geometry states that, if a compact Hermitian manifold has constant holomorphic sectional curvature, then the metric must be Kähler (when the constant is non-zero) or Chern flat (when the constant is zero). It is known to be true in complex dimension $2$ by the work of Balas and Gauduchon in 1985 (when the constant is negative or zero) and Apostolov, Davidov and Muskarov in 1996 (in general). In dimension $3$ or higher, the conjecture is only known in some special cases, such as for all twistor spaces by the work of Davidov, Grantcharov, and Muskarov, or for the locally conformally Kähler case (when the constant is negative or zero) by the work of H. Chen, L. Chen and Nie, or for all non-balanced Bismut torsion parallel (BTP) manifolds by the work of S. Chen and Zheng, where they also showed that the conjecture holds for all balanced BTP threefolds, utilizing a classification result by Zhao and Zheng. In this article, we show that the conjecture is valid for all balanced BTP manifolds in complex dimension $4$. The interesting part is that while balanced BTP manifolds are highly restrictive, the classification is still lacking in dimensions $4$ or higher, and this study might shed some light on the structure of balanced BTP manifolds in dimension $4$.

math.DG↗

Orthogonal complex structures on flat tori

We classify compact Hermitian manifolds with flat Levi-Civita connection. This is equivalent to the classification of all orthogonal complex structures on flat tori. It generalized the work of Khan, Yang, and Zheng in 2017 where they solved the case in complex dimension three.

math.DG↗

Decoupled Structure-Feature Alignment via Alternating Optimization for Graph Learning

Conventional Graph Neural Networks (GNNs) couple feature transformation and neighborhood aggregation, which often renders them vulnerable to topological noise and heterophilous connections. To decouple this dependency, we present a constrained two-view learning framework for robust graph learning, which aligns structure-aware GNN embeddings with a structure-free feature prior. Specifically, the proposed Decoupled Structure-Feature Alternating Learning (DSAL) framework trains an independent anchor network using a self-supervised reconstruction objective to capture the intrinsic semantic information contained in node attributes. Within DSAL, to effectively integrate this prior, we design a channel-split adaptive gated (CSAG) layer. This architecture employs a gating mechanism to balance global spectral smoothing and local spatial representation dynamically. Furthermore, the model is optimized via a cyclic alternating procedure, which mitigates representation drift caused by mutual interference in standard joint optimization schemes. Experiments on diverse homophilous and heterophilous datasets suggest that our proposed approach provides improved node classification accuracy while maintaining robustness to structural perturbations compared to standard message-passing architectures.

cs.LG↗

Emergent Vibronic Spectral Hierarchy in a Kagome Flat-Band Insulator

Electron-phonon coupling is usually understood in terms of electronic quasiparticles interacting with dispersive lattice vibrations. Much less is known about the complementary limit in which the relevant phonon mode is itself localized or weakly dispersive. Here we investigate this regime in the kagome compound Rb$_{2}$Ni$_{3}$S$_{4}$, which undergoes an unconventional insulating transition near $T^{*} \approx$ 260-280~K. Combining polarization-resolved Raman spectroscopy with temperature-dependent x-ray diffraction, scanning tunneling microscopy, and electrical, thermal, and magnetic measurements, we show that the transition involves electronic localization without a conventional structural or magnetic order parameter. Raman spectra reveal a giant Franck-Condon progression associated with a nearly dispersionless 333.7~cm$^{-1}$ phonon, decorated by an equally spaced comb-like fine structure with a characteristic spacing of 40.6~cm$^{-1}$. The comb spacing is insensitive to magnetic field, whereas its spectral weight is strongly field tunable. Rather than treating either hierarchy alone as pure phonon effect, we interpret their nested coexistence as evidence for a strongly coupled electron-vibrational manifold involving a localized lattice coordinate. These results identify dispersionless phonons as an active route to vibronic correlations in solids and suggest that such electron-vibrational self-trapping is closely associated with the insulating phase of Rb$_{2}$Ni$_{3}$S$_{4}$.

cond-mat.str-el↗

Vietoris--Rips Contractibility for Dense Subsets of Finite-Dimensional Normed Spaces

Let $Y$ be a $δ$-dense subset of a finite-dimensional normed space $X$, so every point of $X$ is at distance at most $δ$ from some point of $Y$. We prove that the open Vietoris--Rips complex $\operatorname{VR}_{<}(Y;r)$ is contractible whenever $r>δ/(1-J_{\mathrm{fin}}(X))$, where $J_{\mathrm{fin}}(X)$ is the normalized finite Jung constant: the supremum of circumradius divided by diameter over finite subsets of positive diameter. We use the canonical isometric embedding of $X$ into a hyperconvex normed space, where $\operatorname{VR}_{<}(Y;r)$ is the nerve of an equal-radius open-ball cover. The finite Jung estimate restricts the cover to a convex tube about $X$ without changing its nerve, while $δ$-density makes the restricted balls cover the tube. Thus the complex is homotopy equivalent to the tube and hence contractible. Applying this result to lattices gives explicit contractibility bounds. In $(\mathbb{Z}^n,\ell_1)$, gaps in the distance spectrum sharpen the bound: $\operatorname{VR}_{<}(\mathbb{Z}^n,\ell_1;r)$ is contractible above $\max\{n,n(n-1)/2\}$, and $\operatorname{VR}_{\le}(\mathbb{Z}^n,\ell_1;r)$ is contractible at or above the same value. This asymptotically halves the leading term of Zaremsky's previous bound. We also give explicit bounds for $(\mathbb{Z}^n,\ell_p)$ for all $1\le p\le\infty$, recovering the optimal bound $r=1$ for $p=\infty$; for fixed $n$, the bounds converge as $p\to\infty$. For a general locally finite $δ$-dense subset $Y\subset X$ whose distance spectrum may not be discrete, a separate closed-cover argument proves that $\operatorname{VR}_{\le}(Y;r)$ is contractible whenever $r\geδ/(1-J_{\mathrm{fin}}(X))$.

math.AT↗

General and Efficient Steering of Diffusion Models

Steering diffusion models toward conditions unseen during training typically requires either retraining with conditional inputs or per-step gradient computations, both of which incur substantial computational overhead. We present Noise-Aligned RFM Steering (NA-RFM), a general recipe for efficiently steering diffusion models without gradient guidance during inference, enabling fast controllable generation. The method combines two offline-computed signals: noise alignment, a high-noise correction from PCA statistics of the target examples and the full data, and Recursive Feature Machine (RFM) activation steering, which learns a target-discriminative direction from labeled forward-process activations. During sampling, noise alignment provides coarse control at high noise, while the RFM direction is reused over intermediate/late timesteps through lightweight activation edits. Experiments on CIFAR-10, ImageNet, CelebA, and fine-grained bird species show improved target accuracy over gradient-based post-hoc guidance baselines, improved FID on the class-guidance benchmarks, and substantial inference speedups. Code: https://github.com/isotrivial/na-rfm.

cs.LG↗

OrderDP: A Theoretically Guaranteed Lossless Dynamic Data Pruning Framework

Data pruning (DP), as an oft-stated strategy to alleviate heavy training burdens, reduces the volume of training samples according to a well-defined pruning method while striving for near-lossless performance. However, existing approaches, which commonly select highly informative samples, can lead to biased gradient estimation compared to full-dataset training. Furthermore, the analysis of this bias and its impact on final performance remains ambiguous. To address these challenges, we propose OrderDP, a plug-and-play framework that aims to obtain stable, unbiased, and near-lossless training acceleration with theoretical guarantees. Specifically, OrderDP first randomly selects a subset and then chooses the top-$q$ samples, where unbiasedness is established with respect to a surrogate loss. This ensures that OrderDP conducts unbiased training in terms of the surrogate objective. We further establish convergence and generalization analyses, elucidating how OrderDP affects optimal performance and enables well-controlled acceleration while ensuring guaranteed final performance. Empirically, we evaluate OrderDP against comprehensive baselines on CIFAR-10, CIFAR-100, and ImageNet-1K, demonstrating competitive accuracy, stable convergence, and exact control -- all with a simpler design and faster runtime, while reducing training cost by over 40%. Delivering both strong performance and computational efficiency, our method serves as a robust and easily adaptable tool for data-efficient learning. The code is publicly available at https://github.com/shengze-xu/OrderDP.

cs.LG↗

G-LoG Bi-filtration for Medical Image Classification

Building practical filtrations on objects to detect topological and geometric features is an important task in the field of Topological Data Analysis (TDA). In this paper, leveraging the ability of the Laplacian of Gaussian operator to enhance the boundaries of medical images, we define the G-LoG (Gaussian-Laplacian of Gaussian) bi-filtration to generate the features more suitable for multi-parameter persistence module. By modeling volumetric images as bounded functions, then we prove the interleaving distance on the persistence modules obtained from our bi-filtrations on the bounded functions is stable with respect to the maximum norm of the bounded functions. Finally, we conduct experiments on the MedMNIST dataset, comparing our bi-filtration against single-parameter filtration and the established deep learning baselines, including Google AutoML Vision, ResNet, AutoKeras and auto-sklearn. Experiments results demonstrate that our bi-filtration significantly outperforms single-parameter filtration. Notably, a simple Multi-Layer Perceptron (MLP) trained on the topological features generated by our bi-filtration achieves performance comparable to complex deep learning models trained on the original dataset.

cs.CV↗

Two Calm Ends and the Wild Middle: A Geometric Picture of Memorization in Diffusion Models

Diffusion models generate high-quality samples but can also memorize training data, raising serious privacy concerns. Understanding the mechanisms governing when memorization versus generalization occurs remains an active area of research. In particular, it is unclear where along the noise schedule memorization is induced, how data geometry influences it, and how phenomena at different noise scales interact. We introduce a geometric framework that partitions the noise schedule into three regimes based on the coverage properties of training data by Gaussian shells and the concentration behavior of the posterior, which we argue are two fundamental objects governing memorization and generalization in diffusion models. This perspective reveals that memorization risk is highly non-uniform across noise levels. We further identify a danger zone at medium noise levels where memorization is most pronounced. In contrast, both the small and large noise regimes resist memorization, but through fundamentally different mechanisms: small noise avoids memorization due to limited training coverage, while large noise exhibits low posterior concentration and admits a provably near linear Gaussian denoising behavior. For the medium noise regime, we identify geometric conditions through which we propose a geometry-informed targeted intervention that mitigates memorization.

cs.LG↗