arXiv subjects

Guanliang Li

Publications and source records attributed to Guanliang Li.

5 recordsLinked to original sources

arXiv2026

Real-Time Structural Detection for Indoor Navigation from 3D LiDAR Using Bird's-Eye-View Images

Efficient structural perception is essential for mapping and autonomous navigation on resource-constrained robots. Existing 3D methods are computationally prohibitive, while traditional 2D geometric approaches lack robustness. This paper presents a lightweight, real-time framework that projects 3D LiDAR data into 2D Bird's-Eye-View (BEV) images to enable efficient detection of structural elements relevant to mapping and navigation. Within this representation, we systematically evaluate several feature extraction strategies, including classical geometric techniques (Hough Transform, RANSAC, and LSD) and a deep learning detector based on YOLO-OBB. The resulting detections are integrated through a spatiotemporal fusion module that improves stability and robustness across consecutive frames. Experiments conducted on a standard mobile robotic platform highlight clear performance trade-offs. Classical methods such as Hough and LSD provide fast responses but exhibit strong sensitivity to noise, with LSD producing excessive segment fragmentation that leads to system congestion. RANSAC offers improved robustness but fails to meet real-time constraints. In contrast, the YOLO-OBB-based approach achieves the best balance between robustness and computational efficiency, maintaining an end-to-end latency (satisfying 10 Hz operation) while effectively filtering cluttered observations in a low-power single-board computer (SBC) without using GPU acceleration. The main contribution of this work is a computationally efficient BEV-based perception pipeline enabling reliable real-time structural detection from 3D LiDAR on resource-constrained robotic platforms that cannot rely on GPU-intensive processing. The source code and pre-trained models are publicly available.

cs.RO↗

arXiv2026

V-FAT: Benchmarking Visual Fidelity Against Text-bias

Recent advancements in Multimodal Large Language Models (MLLMs) have demonstrated impressive performance on standard visual reasoning benchmarks. However, there is growing concern that these models rely excessively on linguistic shortcuts rather than genuine visual grounding, a phenomenon we term Text Bias. In this paper, we investigate the fundamental tension between visual perception and linguistic priors. We decouple the sources of this bias into two dimensions: Internal Corpus Bias, stemming from statistical correlations in pretraining, and External Instruction Bias, arising from the alignment-induced tendency toward sycophancy. To quantify this effect, we introduce V-FAT (Visual Fidelity Against Text-bias), a diagnostic benchmark comprising 4,026 VQA instances across six semantic domains. V-FAT employs a Three-Level Evaluation Framework that systematically increases the conflict between visual evidence and textual information: (L1) internal bias from atypical images, (L2) external bias from misleading instructions, and (L3) synergistic bias where both coincide. We introduce the Visual Robustness Score (VRS), a metric designed to penalize "lucky" linguistic guesses and reward true visual fidelity. Our evaluation of 12 frontier MLLMs reveals that while models excel in existing benchmarks, they experience significant visual collapse under high linguistic dominance.

cs.CL↗

arXiv2025

Learning High-Order Relationships with Hypergraph Attention-based Spatio-Temporal Aggregation for Brain Disease Analysis

Traditional functional connectivity based on functional magnetic resonance imaging (fMRI) can only capture pairwise interactions between brain regions. Hypergraphs, which reveal high-order relationships among multiple brain regions, have been widely used for disease analysis. However, existing methods often rely on predefined hypergraph structures, limiting their ability to model complex patterns. Moreover, temporal information, an essential component of brain high-order relationships, is frequently overlooked. To address these limitations, we propose a novel framework that jointly learns informative and sparse high-order brain structures along with their temporal dynamics. Inspired by the information bottleneck principle, we introduce an objective that maximizes information and minimizes redundancy, aiming to retain disease-relevant high-order features while suppressing irrelevant signals. Our model comprises a multi-hyperedge binary mask module for hypergraph structure learning, a hypergraph self-attention aggregation module that captures spatial features through adaptive attention across nodes and hyperedges, and a spatio-temporal low-dimensional network for extracting discriminative spatio-temporal representations for disease classification. Experiments on benchmark fMRI datasets demonstrate that our method outperforms the state-of-the-art approaches and successfully identifies meaningful high-order brain interactions. These findings provide new insights into brain network modeling and the study of neuropsychiatric disorders.

q-bio.NC↗

arXiv2013

An adaptive GMsFEM for high-contrast flow problems

In this paper, we derive an a-posteriori error indicator for the Generalized Multiscale Finite Element Method (GMsFEM) framework. This error indicator is further used to develop an adaptive enrichment algorithm for the linear elliptic equation with multiscale high-contrast coefficients. The GMsFEM, which has recently been introduced in [12], allows solving multiscale parameter-dependent problems at a reduced computational cost by constructing a reduced-order representation of the solution on a coarse grid. The main idea of the method consists of (1) the construction of snapshot space, (2) the construction of the offline space, and (3) the construction of the online space (the latter for parameter-dependent problems). In [12], it was shown that the GMsFEM provides a flexible tool to solve multiscale problems with a complex input space by generating appropriate snapshot, offline, and online spaces. In this paper, we study an adaptive enrichment procedure and derive an a-posteriori error indicator which gives an estimate of the local error over coarse grid regions. We consider two kinds of error indicators where one is based on the $L^2$-norm of the local residual and the other is based on the weighted $H^{-1}$-norm of the local residual where the weight is related to the coefficient of the elliptic equation. We show that the use of weighted $H^{-1}$-norm residual gives a more robust error indicator which works well for cases with high contrast media. The convergence analysis of the method is given. In our analysis, we do not consider the error due to the fine-grid discretization of local problems and only study the errors due to the enrichment. Numerical results are presented that demonstrate the robustness of the proposed error indicators.

math.NA↗

arXiv2007

Transport and Percolation Theory in Weighted Networks

We study the distribution $P(σ)$ of the equivalent conductance $σ$ for Erdős-Rényi (ER) and scale-free (SF) weighted resistor networks with $N$ nodes. Each link has conductance $g\equiv e^{-ax}$, where $x$ is a random number taken from a uniform distribution between 0 and 1 and the parameter $a$ represents the strength of the disorder. We provide an iterative fast algorithm to obtain $P(σ)$ and compare it with the traditional algorithm of solving Kirchhoff equations. We find, both analytically and numerically, that $P(σ)$ for ER networks exhibits two regimes. (i) A low conductance regime for $σ< e^{-ap_c}$ where $p_c=1/\av{k}$ is the critical percolation threshold of the network and $\av{k}$ is average degree of the network. In this regime $P(σ)$ is independent of $N$ and follows the power law $P(σ) \sim σ^{-α}$, where $α=1-\av{k}/a$. (ii) A high conductance regime for $σ>e^{-ap_c}$ in which we find that $P(σ)$ has strong $N$ dependence and scales as $P(σ) \sim f(σ,ap_c/N^{1/3})$. For SF networks with degree distribution $P(k)\sim k^{-λ}$, $k_{min} \le k \le k_{max}$, we find numerically also two regimes, similar to those found for ER networks.

cond-mat.stat-mech↗