arXiv subjects

Yinghan Zhang

Publications and source records attributed to Yinghan Zhang.

4 recordsLinked to original sources

arXiv2026

DobotTrack: Target-Grounded Vision-Language-Action Learning for Embodied Person Tracking

Language-guided embodied person tracking requires a mobile agent to identify a target from a referring description and consistently follow the target under egocentric observations. We present DobotTrack, a target-grounded vision-language-action (VLA) framework that integrates auxiliary grounding supervision with efficient temporal modeling. We systematically investigate five auxiliary tasks---person segmentation, bounding-box grounding, visual-spatial classification (VSC), future DINO feature prediction, and future-frame prediction---and adopt complementary mask and bounding-box supervision to enhance target delineation and localization. We introduce Q-identity, a target-conditioned historical visual compression module that leverages target-grounding predictions to selectively preserve identity-discriminative appearance features and spatial cues in compact historical tokens. On EVT-Bench, our 0.8B-parameter model achieves an average success rate of 77.9%, exceeding OmTrackVLA by 16.9 percentage points across the three tasks. We further construct a real-world person-tracking dataset collected using a quadruped robot, comprising 126 episodes and 596,323 mask-annotated frames. The target masks enable automatic derivation of bounding boxes and visual-spatial labels encoding horizontal position and apparent image scale, providing additional grounding supervision without separate manual annotation of these labels. Together, our framework and annotation pipeline support learning target-grounded tracking policies from simulated and real-world data. Code is available at https://github.com/Dobot-Team/dobot_track.

cs.RO↗

arXiv2026

Stopping Rules and the Poisson Boundary of Random Walks on Countable Groups

In this paper, we study the terminal laws and Poisson boundaries obtained by stopping random walks on countable groups. For a transient walk, we give a direct Green-potential characterization of randomized stopping distributions and show that the terminal law determines the mean stopping time. On free groups, this yields finitely supported randomized stopping distributions at a positive total variation distance from every ordinary stopping distribution. We construct symmetric, fully supported recurrent measures of finite entropy on countable locally finite groups and obtain ordinary strict boundary enlargements when the group is not Choquet-Deny. For generating source and terminal measures of finite entropy, positive asymptotic entropy of the source forces the stopping time to have finite mean. A direct proof of the finite-mean entropy formula and boundary preservation then shows that a finite-entropy stopping chain has at most one strict enlargement. An exact stopped-word entropy identity shows that infinite-mean ordinary stopping with a finite-entropy terminal law has infinite conditional history entropy.

math.PR↗

arXiv2025

On the Dirichlet Problem at Infinity and Poisson Boundary for Certain Manifolds without Conjugate Points

In this paper, we investigate the problem of the existence of the bounded harmonic functions on a simply connected Riemannian manifold $\widetilde{M}$ without conjugate points, which can be compactified via the ideal boundary $\widetilde{M}(\infty)$. Let $\widetilde{M}$ be a uniform visibility manifold which satisfy the Axiom $2$, or a rank $1$ manifold without focal points, suppose that $Γ$ is a cocompact discrete subgroup of $Iso(\widetilde{M})$, we show that for a given continuous function on $\widetilde{M}(\infty)$, there exists a harmonic extension to $\widetilde{M}$. And furthermore, when $\widetilde{M}$ is a rank $1$ manifold without focal points, the Brownian motion defines a family of harmonic measures $ν_{\ast}$ on $\widetilde{M}(\infty)$, we show that $(\widetilde{M}(\infty),ν_{\ast})$ is isomorphic to the Poisson boundary of $Γ$.

math.DG↗

arXiv2025

Trustworthy Enhanced Multi-view Multi-modal Alzheimer's Disease Prediction with Brain-wide Imaging Transcriptomics Data

Brain transcriptomics provides insights into the molecular mechanisms by which the brain coordinates its functions and processes. However, existing multimodal methods for predicting Alzheimer's disease (AD) primarily rely on imaging and sometimes genetic data, often neglecting the transcriptomic basis of brain. Furthermore, while striving to integrate complementary information between modalities, most studies overlook the informativeness disparities between modalities. Here, we propose TMM, a trusted multiview multimodal graph attention framework for AD diagnosis, using extensive brain-wide transcriptomics and imaging data. First, we construct view-specific brain regional co-function networks (RRIs) from transcriptomics and multimodal radiomics data to incorporate interaction information from both biomolecular and imaging perspectives. Next, we apply graph attention (GAT) processing to each RRI network to produce graph embeddings and employ cross-modal attention to fuse transcriptomics-derived embedding with each imagingderived embedding. Finally, a novel true-false-harmonized class probability (TFCP) strategy is designed to assess and adaptively adjust the prediction confidence of each modality for AD diagnosis. We evaluate TMM using the AHBA database with brain-wide transcriptomics data and the ADNI database with three imaging modalities (AV45-PET, FDG-PET, and VBM-MRI). The results demonstrate the superiority of our method in identifying AD, EMCI, and LMCI compared to state-of-the-arts. Code and data are available at https://github.com/Yaolab-fantastic/TMM.

cs.AI↗