arXiv subjects

Xiaoyu He

Publications and source records attributed to Xiaoyu He.

At least 19 recordsLinked to original sources

arXiv2026

MoCapAnything V2: End-to-End Motion Capture for Arbitrary Skeletons

Recent methods for arbitrary-skeleton motion capture from monocular video follow a factorized pipeline, where a Video-to-Pose network predicts joint positions and an analytical inverse-kinematics (IK) stage recovers joint rotations. While effective, this design is inherently limited, since joint positions do not fully determine rotations and leave degrees of freedom such as bone-axis twist ambiguous, and the non-differentiable IK stage prevents the system from adapting to noisy predictions or optimizing for the final animation objective. In this work, we present the first fully end-to-end framework in which both Video-to-Pose and Pose-to-Rotation are learnable and jointly optimized. We observe that the ambiguity in pose-to-rotation mapping arises from missing coordinate system information: the same joint positions can correspond to different rotations under different rest poses and local axis conventions. To resolve this, we introduce a reference pose-rotation pair from the target asset, which, together with the rest pose, not only anchors the mapping but also defines the underlying rotation coordinate system. This formulation turns rotation prediction into a well-constrained conditional problem and enables effective learning. In addition, our model predicts joint positions directly from video without relying on mesh intermediates, improving both robustness and efficiency. Both stages share a skeleton-aware Global-Local Graph-guided Multi-Head Attention (GL-GMHA) module for joint-level local reasoning and global coordination. Experiments on Truebones Zoo and Objaverse show that our method reduces rotation error from ~17 degrees to ~10 degrees, and to 6.54 degrees on unseen skeletons, while achieving ~20x faster inference than mesh-based pipelines. Project page: https://animotionlab.github.io/MoCapAnythingV2/

cs.CV↗

arXiv2026

Logarithmically larger deletion codes of all distances

The deletion distance between two binary words $u,v \in \{0,1\}^n$ is the smallest $k$ such that $u$ and $v$ share a common subsequence of length $n-k$. A set $C$ of binary words of length $n$ is called a $k$-deletion code if every pair of distinct words in $C$ has deletion distance greater than $k$. In 1965, Levenshtein initiated the study of deletion codes by showing that, for $k\ge 1$ fixed and $n$ going to infinity, a $k$-deletion code $C\subseteq \{0,1\}^n$ of maximum size satisfies $Ω_k(2^n/n^{2k}) \leq |C| \leq O_k( 2^n/n^k)$. We make the first asymptotic improvement to these bounds by showing that there exist $k$-deletion codes with size at least $Ω_k(2^n \log n/n^{2k})$. Our proof is inspired by Jiang and Vardy's improvement to the classical Gilbert--Varshamov bounds. We also establish several related results on the number of longest common subsequences and shortest common supersequences of a pair of words with given length and deletion distance.

math.CO↗

arXiv2026

Decentralized Federated Learning for Heterogeneous Multi-Task Semantic Communication

Collaborative training in distributed semantic communication (DSC) networks typically relies on decentralized federated learning (DFL). However, pushing topology-agnostic aggregation into heterogeneous, multi-task environments creates a fundamental bottleneck: it drives negative transfer and overconsensus bias (OCB). This paper introduces a personalized DSC framework that cuts off this cross-task interference. At the node level, a policy-driven multi-path routing mechanism separates task-specific features from shared representations to preserve local fidelity. Across the network, we deploy a "communicationwhile- aggregation" protocol. It calibrates a column-stochastic consensus matrix using task affinities. This limits the system to absorbing complementary knowledge while actively blocking mismatched parameter updates. To bound the convergence, we derive a unified Lyapunov drift analysis. We reveal a strict Ushaped trade-off: deeper topological mixing reduces variance but amplifies structural OCB. Resolving this tension yields a closed-form expression for the optimal aggregation depth. We evaluate the proposed framework on NYU-v2, where the results reveal a clear trade-off between insufficient aggregation and excessive topological mixing. At the analytically derived optimal aggregation depth, our method achieves a 4.77% global relative improvement over the no-aggregation baseline and outperforms decentralized FedAvg, FedAMP, and heuristic max aggregation. We further evaluate the framework on Taskonomy and imperfect wireless links to examine the effects of network-size variation and wireless-link reliability.

cs.AI↗

arXiv2026

A reformulation of the discrete Convexity Conjecture via $k$-thresholds

We introduce the notion of "$k$-thresholds'' and show that Talagrand's discrete convexity conjecture is equivalent to the assertion that, for some universal integer $k \ge 2$, the $k$-threshold of every increasing family is at most a universal constant times its expectation threshold. We prove a reduction theorem that bounds the $k$-threshold of any increasing graph property in terms of ordinary thresholds of graphs in suitable decompositions of its members. As a consequence, we determine, up to a constant factor, the $k$-threshold of every fixed graph in terms of a natural $k$-density parameter. We also prove that $k=2$ suffices for several classical spanning graph containment properties. More generally, we establish the conjectured comparison between $k$-thresholds and expectation thresholds for broad classes of graph containment properties whose target graphs have low degeneracy.

math.CO↗

arXiv2026

ExMAG: Learning of Maximally Ancestral Graphs

In mixed graphs, there are both directed and bidirected edges. An extension of acyclicity to this mixed-graph setting is known as maximally ancestral graphs. This extension is of considerable interest in causal learning in the presence of confounders. There, directed edges represent a clear direction of causality, while bidirected edges represent confounding. We propose a branch-and-cut algorithm for learning maximally ancestral graphs using a formulation as a mixed-integer quadratic program. Empirically, our method achieves comparable or improved reconstruction quality while requiring an order of magnitude fewer samples than state-of-the-art approaches.

cs.LG↗

arXiv2026

AUSLUN: A Fixed-Hover UAV--USV System for GNSS-Denied Maritime Search and Navigation

Global navigation satellite system (GNSS) denial can prevent an unmanned surface vehicle (USV) from both finding a distant vessel and maintaining a globally referenced approach. This paper presents AUSLUN (Automatic UAV Search, Localization, and USV Navigation), a fixed-hover aerial-surface system that uses a coastal unmanned aerial vehicle (UAV), which estimates its own pose through visual-inertial odometry (VIO), as a long-range sensing and navigation anchor. The central design shifts sensing motion from UAV translation to a zoom pod and closes the loop through three coupled elements: polygon-aware annular pod scanning, modality-aware bearing-range localization, and target-relative USV guidance with visual-loss recovery. The same gated recursive estimator uses laser range for the non-cooperative target and datalink range for the cooperative USV. Search-planning simulations show that the adaptive yaw bounds reduce scan time and redundant coverage relative to a matched fixed-sector scan, and GPS-referenced field data show that the gated recursive estimator outperforms non-recursive baselines in localization accuracy. An integrated maritime mission further demonstrates the complete search-to-navigation sequence, including a deliberately triggered visual-loss recovery. These results establish the feasibility and operating boundary of fixed-hover UAV assistance for stationary-target approach in coastal GNSS-denied environments. The source code and a video demonstration are publicly available at https://github.com/xirhxq/pod_search and https://youtu.be/S-5RkJs35JI.

cs.RO↗

arXiv2026

MoCapAnything: Unified 3D Motion Capture for Arbitrary Skeletons from Monocular Videos

Motion capture now underpins content creation far beyond digital humans, yet most existing pipelines remain species- or template-specific. We formalize this gap as Category-Agnostic Motion Capture (CAMoCap): given a monocular video and an arbitrary rigged 3D asset as a prompt, the goal is to reconstruct a rotation-based animation such as BVH that directly drives the specific asset. We present MoCapAnything, a reference-guided, factorized framework that first predicts 3D joint trajectories and then recovers asset-specific rotations via constraint-aware inverse kinematics. The system contains three learnable modules and a lightweight IK stage: (1) a Reference Prompt Encoder that extracts per-joint queries from the asset's skeleton, mesh, and rendered images; (2) a Video Feature Extractor that computes dense visual descriptors and reconstructs a coarse 4D deforming mesh to bridge the gap between video and joint space; and (3) a Unified Motion Decoder that fuses these cues to produce temporally coherent trajectories. We also curate Truebones Zoo with 1038 motion clips, each providing a standardized skeleton-mesh-render triad. Experiments on both in-domain benchmarks and in-the-wild videos show that MoCapAnything delivers high-quality skeletal animations and exhibits meaningful cross-species retargeting across heterogeneous rigs, enabling scalable, prompt-driven 3D motion capture for arbitrary assets. Project page: https://animotionlab.github.io/MoCapAnything/

cs.CV↗

arXiv2026

Causal Learning in Biomedical Applications: Krebs Cycle as a Benchmark

Learning causal relationships from time series data is an important but challenging problem. Existing synthetic datasets often contain hidden artifacts that can be exploited by causal discovery methods, reducing their usefulness for benchmarking. We present a new benchmark dataset based on simulations of the Krebs cycle, a key biochemical pathway. The data are generated using a particle-based simulator that models molecular interactions in a controlled environment. Four distinct scenarios are provided, varying in time series length, number of samples, and intervention settings. The benchmark includes ground-truth causal graphs for evaluation. It supports quantitative comparisons using metrics such as Structural Hamming Distance, Structural Intervention Distance, and F1-score. A comprehensive evaluation of 14 causal discovery methods from different modelling paradigms is presented. Performance is compared across datasets using multiple accuracy and efficiency measures. The dataset provides a reproducible, interpretable, and non-trivial benchmark for testing causal learning methods on time series data. It avoids common pitfalls such as residual structural patterns and supports interventions and evaluation with known causal ground truth. This makes it a useful tool for the development and comparison of causal discovery algorithms.

cs.LG↗

arXiv2026

Hypergraph Ramsey numbers with quasipolynomial growth rate

For a 3-uniform hypergraph (3-graph) $F$, let $r(F,n)$ be the smallest $N$ such that any $N$-vertex $F$-free 3-graph has an independent set of size $n$. We construct a $3$-graph $H_2$ with six vertices and five edges such that $r(H_2,n)=n^{Θ(\log n)}$, and a more general family of $3$-graphs $F$ for which $r(F,n)=n^{\log^{Θ(1)}(n)}$. These are the first examples of such Ramsey number known to be neither polynomial nor exponential.

math.CO↗

arXiv2026

Joint Routing and Model Pruning for Decentralized Federated Learning in Bandwidth-Constrained Multi-Hop Wireless Networks

Decentralized federated learning (D-FL) enables privacy-preserving training without a central server, but multi-hop model exchanges and aggregation are often bottlenecked by communication resource constraints. To address this issue, we propose a joint routing-and-pruning framework that optimizes routing paths and pruning rates to maintain communication latency within prescribed limits. We analyze how the sum of model biases across all clients affects the convergence bound of D-FL and formulate an optimization problem that maximizes the model retention rate to minimize these biases under communication constraints. Further analysis reveals that each client's model retention rate is path-dependent, which reduces the original problem to a routing optimization. Leveraging this insight, we develop a routing algorithm that selects latency-efficient transmission paths, allowing more parameters to be delivered within the time budget and thereby improving D-FL convergence. Simulations demonstrate that, compared with unpruned systems, the proposed framework reduces average transmission latency by 27.8% and improves testing accuracy by approximately 12%. Furthermore, relative to standard benchmark routing algorithms, the proposed routing method improves accuracy by roughly 8%.

cs.LG↗

arXiv2026

Projection-Free Evolution Strategies for Continuous Prompt Search

Continuous prompt search offers a computationally efficient alternative to conventional parameter tuning in natural language processing tasks. Nevertheless, its practical effectiveness can be significantly hindered by the black-box nature and the inherent high-dimensionality of the objective landscapes. Existing methods typically mitigate these challenges by restricting the search to a randomly projected low-dimensional subspace. However, the effectiveness and underlying motivation of the projection mechanism remain ambiguous. In this paper, we first empirically demonstrate that despite the prompt space possessing a low-dimensional structure, random projections fail to adequately capture this essential structure. Motivated by this finding, we propose a projection-free prompt search method based on evolutionary strategies. By directly optimizing in the full prompt space with an adaptation mechanism calibrated to the intrinsic dimension, our method achieves competitive search capabilities without additional computational overhead. Furthermore, to bridge the generalization gap in few-shot scenarios, we introduce a confidence-based regularization mechanism that systematically enhances the model's confidence in the target verbalizers. Experimental results on seven natural language understanding tasks from the GLUE benchmark demonstrate that our proposed approach significantly outperforms existing baselines.

cs.CL↗

arXiv2026

Rational values of the weak saturation limit

Given a graph $F$, a graph $G$ is weakly $F$-saturated if all non-edges of $G$ can be added in some order so that each new edge introduces a copy of $F$. The weak saturation number $\operatorname{wsat}(n, F)$ is the minimum number of edges in a weakly $F$-saturated graph on $n$ vertices. Bollobás initiated the study of weak saturation in 1968 to study percolation processes, which originated in biology and have applications in physics and computer science. It was shown by Alon that for each $F$, there is a constant $w_F$ such that $\operatorname{wsat}(n, F) = w_Fn + o(n)$. We characterize all possible rational values of $w_F$, proving in particular that $w_F$ can equal any rational number at least $\frac 32$.

math.CO↗

arXiv2026

DiMo: Discrete Diffusion Modeling for Motion Generation and Understanding

Prior masked modeling motion generation methods predominantly study text-to-motion. We present DiMo, a discrete diffusion-style framework, which extends masked modeling to bidirectional text--motion understanding and generation. Unlike GPT-style autoregressive approaches that tokenize motion and decode sequentially, DiMo performs iterative masked token refinement, unifying Text-to-Motion (T2M), Motion-to-Text (M2T), and text-free Motion-to-Motion (M2M) within a single model. This decoding paradigm naturally enables a quality-latency trade-off at inference via the number of refinement steps. We further improve motion token fidelity with residual vector quantization (RVQ) and enhance alignment and controllability with Group Relative Policy Optimization (GRPO). Experiments on HumanML3D and KIT-ML show strong motion quality and competitive bidirectional understanding under a unified framework. In addition, we demonstrate model ability in text-free motion completion, text-guided motion prediction and motion caption correction without architectural change. Additional qualitative results are available on our project page: https://animotionlab.github.io/DiMo/.

cs.CV↗

arXiv2026

NECromancer: Breathing Life into Skeletons via BVH Animation

Motion tokenization is a key component of generalizable motion models, yet most existing approaches are restricted to species-specific skeletons, limiting their applicability across diverse morphologies. We propose NECromancer (NEC), a universal motion tokenizer that operates directly on arbitrary BVH skeletons. NEC consists of three components: (1) an Ontology-aware Skeletal Graph Encoder (OwO) that encodes structural priors from BVH files, including joint semantics, rest-pose offsets, and skeletal topology, into skeletal embeddings; (2) a Topology-Agnostic Tokenizer (TAT) that compresses motion sequences into a universal, topology-invariant discrete representation; and (3) the Unified BVH Universe (UvU), a large-scale dataset aggregating BVH motions across heterogeneous skeletons. Experiments show that NEC achieves high-fidelity reconstruction under substantial compression and effectively disentangles motion from skeletal structure. The resulting token space supports cross-species motion transfer, composition, denoising, generation with token-based models, and text-motion retrieval, establishing a unified framework for motion analysis and synthesis across diverse morphologies. Demo page: https://animotionlab.github.io/NECromancer/

cs.CV↗

arXiv2026

A Provable Expressiveness Hierarchy in Hybrid Linear-Full Attention

Transformers serve as the foundation of most modern large language models. To mitigate the quadratic complexity of standard full attention, various efficient attention mechanisms, such as linear and hybrid attention, have been developed. A fundamental gap remains: their expressive power relative to full attention lacks a rigorous theoretical characterization. In this work, we theoretically characterize the performance differences among these attention mechanisms. Our theory applies to all linear attention variants that can be formulated as a recurrence, including Mamba, DeltaNet, etc. Specifically, we establish an expressiveness hierarchy: for the sequential function composition-a multi-step reasoning task that must occur within a model's forward pass, an ($L+1$)-layer full attention network is sufficient, whereas any hybrid network interleaving $L-1$ layers of full attention with a substantially larger number ($2^{3L^2}$) of linear attention layers cannot solve it. This result demonstrates a clear separation in expressive power between the two types of attention. Our work provides the first provable separation between hybrid attention and standard full attention, offering a theoretical perspective for understanding the fundamental capabilities and limitations of different attention mechanisms.

cs.LG↗

arXiv2026

BAPS: A Fine-Grained Low-Precision Scheme for Softmax in Attention via Block-Aware Precision reScaling

As the performance gains from accelerating quantized matrix multiplication plateau, the softmax operation becomes the critical bottleneck in Transformer inference. This bottleneck stems from two hardware limitations: (1) limited data bandwidth between matrix and vector compute cores, and (2) the significant area cost of high-precision (FP32/16) exponentiation units (EXP2). To address these issues, we introduce a novel low-precision workflow that employs a specific 8-bit floating-point format (HiF8) and block-aware precision rescaling for softmax. Crucially, our algorithmic innovations make low-precision softmax feasible without the significant model accuracy loss that hampers direct low-precision approaches. Specifically, our design (i) halves the required data movement bandwidth by enabling matrix multiplication outputs constrained to 8-bit, and (ii) substantially reduces the EXP2 unit area by computing exponentiations in low (8-bit) precision. Extensive evaluation on language models and multi-modal models confirms the validity of our method. By alleviating the vector computation bottleneck, our work paves the way for doubling end-to-end inference throughput without increasing chip area, and offers a concrete co-design path for future low-precision hardware and software.

cs.LG↗

arXiv2025

A question of Erdős and Graham on Egyptian fractions

Answering a question of Erdős and Graham, we show that for each fixed positive rational number $x$ the number of ways to write $x$ as a sum of reciprocals of distinct positive integers each at most $n$ is $2^{(c_x + o(1))n}$ for an explicit constant $c_x$ increasing with $x$.

math.CO↗

arXiv2025

Alada: Alternating Adaptation of Momentum Method for Memory-Efficient Matrix Optimization

This work proposes Alada, an adaptive momentum method for stochastic optimization over large-scale matrices. Alada employs a rank-one factorization approach to estimate the second moment of gradients, where factors are updated alternatively to minimize the estimation error. Alada achieves sublinear memory overheads and can be readily extended to optimizing tensor-shaped variables.We also equip Alada with a first moment estimation rule, which enhances the algorithm's robustness without incurring additional memory overheads. The theoretical performance of Alada aligns with that of traditional methods such as Adam. Numerical studies conducted on several natural language processing tasks demonstrate the reduction in memory overheads and the robustness in training large models relative to Adam and its variants.

cs.LG↗