arXiv subjects

Zhiwei Huang

Publications and source records attributed to Zhiwei Huang.

At least 19 recordsLinked to original sources

arXiv2026

SCC-Loc: A Unified Semantic Cascade Consensus Framework for UAV Thermal Geo-Localization

Cross-modal Thermal Geo-localization (TG) provides a robust, all-weather solution for Unmanned Aerial Vehicles (UAVs) in Global Navigation Satellite System (GNSS)-denied environments. However, profound thermal-visible modality gaps introduce severe feature ambiguity, systematically corrupting conventional coarse-to-fine registration. To dismantle this bottleneck, we propose SCC-Loc, a unified Semantic-Cascade-Consensus localization framework. By sharing a single DINOv2 backbone across global retrieval and MINIMA$_{\text{RoMa}}$ matching, it minimizes memory footprint and achieves zero-shot, highly accurate absolute position estimation. Specifically, we tackle modality ambiguity by introducing three cohesive components. First, we design the Semantic-Guided Viewport Alignment (SGVA) module to adaptively optimize satellite crop regions, effectively correcting initial spatial deviations. Second, we develop the Cascaded Spatial-Adaptive Texture-Structure Filtering (C-SATSF) mechanism to explicitly enforce geometric consistency, thereby eradicating dense cross-modal outliers. Finally, we propose the Consensus-Driven Reliability-Aware Position Selection (CD-RAPS) strategy to derive the optimal solution through a synergy of physically constrained pose optimization. To address data scarcity, we construct Thermal-UAV, a comprehensive dataset providing 11,890 diverse thermal queries referenced against a large-scale satellite ortho-photo and corresponding spatially aligned Digital Surface Model (DSM). Extensive experiments demonstrate that SCC-Loc establishes a new state-of-the-art, suppressing the mean localization error to 9.37 m and providing a 7.6-fold accuracy improvement within a strict 5-m threshold over the strongest baseline. Code and dataset are available at https://github.com/FloralHercules/SCC-Loc.

cs.CV↗

arXiv2026

Unveiling the impact of anti-site defects in magnetic transitions of few-layer MnBi2Te4 by operando heating

As the first experimentally discovered intrinsic magnetic topological insulator, MnBi2Te4 has attracted widespread attentions, providing a unique platform for the exploration of topological quantum phases, such as quantum anomalous Hall effect and axion insulator state. Despite the increasing number of potential factors affecting samples being identified, obtaining the high-quality device performance with desired topological quantum phases remains a challenge. In this work, by comparing the reflective magnetic circular dichroism (RMCD) of crystals with different defect densities that are characterized by atomically resolved scanning tunneling microscopy, we demonstrate that anti-site defects play an essential role in achieving ideal magnetic states. By measuring RMCD hysteresis loops with operando heating, we find that MnBi2Te4 few-layer samples are highly susceptible to thermal impact, even at temperature as low as 45°C. The magnetic behavior of heating-treated samples is akin to that of samples fabricated into devices, revealing the thermal impact on devices as well. Starting from few-layers with ideal layer-dependent magnetic order, thermal heating leads to the convergence of magnetization and transition fields between odd- and even-layers. The observed heating-induced magnetic evolution can serve as a valuable reference for assessing the sample quality or the density of anti-site defects. Our findings not only point out the long-standing hidden factor that arose controversies in MnBi2Te4, but also pave the way for controllably engineering the topological quantum phenomena.

cond-mat.mtrl-sci↗

arXiv2026

DST-Calib: A Dual-Path, Self-Supervised, Target-Free LiDAR-Camera Extrinsic Calibration Network

LiDAR-camera extrinsic calibration is essential for multi-modal data fusion in robotic perception systems. However, existing approaches typically rely on handcrafted calibration targets (e.g., checkerboards) or specific, static scene types, limiting their adaptability and deployment in real-world autonomous and robotic applications. This article presents the first self-supervised LiDAR-camera extrinsic calibration network that operates in an online fashion and eliminates the need for specific calibration targets. We first identify a significant generalization degradation problem in prior methods, caused by the conventional single-sided data augmentation strategy. To overcome this limitation, we propose a novel double-sided data augmentation technique that generates multi-perspective camera views using estimated depth maps, thereby enhancing robustness and diversity during training. Built upon this augmentation strategy, we design a dual-path, self-supervised calibration framework that reduces the dependence on high-precision ground truth labels and supports fully adaptive online calibration. Furthermore, to improve cross-modal feature association, we replace the traditional dual-branch feature extraction design with a difference map construction process that explicitly correlates LiDAR and camera features. This not only enhances calibration accuracy but also reduces model complexity. Extensive experiments conducted on five public benchmark datasets, as well as our own recorded dataset, demonstrate that the proposed method significantly outperforms existing approaches in terms of generalizability.

cs.RO↗

arXiv2025

Environment-Driven Online LiDAR-Camera Extrinsic Calibration

LiDAR-camera extrinsic calibration (LCEC) is crucial for multi-modal data fusion in autonomous robotic systems. Existing methods, whether target-based or target-free, typically rely on customized calibration targets or fixed scene types, which limit their applicability in real-world scenarios. To address these challenges, we present EdO-LCEC, the first environment-driven online calibration approach. Unlike traditional target-free methods, EdO-LCEC employs a generalizable scene discriminator to estimate the feature density of the application environment. Guided by this feature density, EdO-LCEC extracts LiDAR intensity and depth features from varying perspectives to achieve higher calibration accuracy. To overcome the challenges of cross-modal feature matching between LiDAR and camera, we introduce dual-path correspondence matching (DPCM), which leverages both structural and textural consistency for reliable 3D-2D correspondences. Furthermore, we formulate the calibration process as a joint optimization problem that integrates global constraints across multiple views and scenes, thereby enhancing overall accuracy. Extensive experiments on real-world datasets demonstrate that EdO-LCEC outperforms state-of-the-art methods, particularly in scenarios involving sparse point clouds or partially overlapping sensor views.

cs.CV↗

arXiv2025

ChangingGrounding: 3D Visual Grounding in Changing Scenes

Real-world robots localize objects from natural-language instructions while scenes around them keep changing. Yet most of the existing 3D visual grounding (3DVG) method still assumes a reconstructed and up-to-date point cloud, an assumption that forces costly re-scans and hinders deployment. We argue that 3DVG should be formulated as an active, memory-driven problem, and we introduce ChangingGrounding, the first benchmark that explicitly measures how well an agent can exploit past observations, explore only where needed, and still deliver precise 3D boxes in changing scenes. To set a strong reference point, we also propose Mem-ChangingGrounder, a zero-shot method for this task that marries cross-modal retrieval with lightweight multi-view fusion: it identifies the object type implied by the query, retrieves relevant memories to guide actions, then explores the target efficiently in the scene, falls back when previous operations are invalid, performs multi-view scanning of the target, and projects the fused evidence from multi-view scans to get accurate object bounding boxes. We evaluate different baselines on ChangingGrounding, and our Mem-ChangingGrounder achieves the highest localization accuracy while greatly reducing exploration cost. We hope this benchmark and method catalyze a shift toward practical, memory-centric 3DVG research for real-world applications. Project page: https://hm123450.github.io/CGB/ .

cs.CV↗

arXiv2025

LiveThinking: Enabling Real-Time Efficient Reasoning for AI-Powered Livestreaming via Reinforcement Learning

In AI-powered e-commerce livestreaming, digital avatars require real-time responses to drive engagement, a task for which high-latency Large Reasoning Models (LRMs) are ill-suited. We introduce LiveThinking, a practical two-stage optimization framework to bridge this gap. First, we address computational cost by distilling a 670B teacher LRM into a lightweight 30B Mixture-of-Experts (MoE) model (3B active) using Rejection Sampling Fine-Tuning (RFT). This reduces deployment overhead but preserves the teacher's verbose reasoning, causing latency. To solve this, our second stage employs reinforcement learning with Group Relative Policy Optimization (GRPO) to compress the model's reasoning path, guided by a multi-objective reward function balancing correctness, helpfulness, and brevity. LiveThinking achieves a 30-fold reduction in computational cost, enabling sub-second latency. In real-world application on Taobao Live, it improved response correctness by 3.3% and helpfulness by 21.8%. Tested by hundreds of thousands of viewers, our system led to a statistically significant increase in Gross Merchandise Volume (GMV), demonstrating its effectiveness in enhancing user experience and commercial performance in live, interactive settings.

cs.LG↗

arXiv2025

Quantized Topological States and Parity Anomaly in Intrinsic Quantum Anomalous Hall Insulator MnBi2Te4

When thinned down to just a few atomic layers, the layered magnetic topological insulator MnBi2Te4 offers an exceptional platform for exploring a wide range of topological phenomena. In this work, we overcome longstanding challenges in synthesizing high-purity MnBi2Te4 crystals and report the observation of a myriad of quantized topological states in high-quality five-septuple-layer (5-SL) samples under magnetic fields up to 45 Tesla. We show that the nontrivial topology of 5-SL MnBi2Te4, in the presence of Landau quantization, is governed by a generalized topological index rooted in the parity anomaly of Dirac fermions in (2+1) dimensions. The anomaly manifests as an anomalous Landau level, giving rise to gate-tunable helical edge transport. Our results establish high-quality MnBi2Te4 as a robust platform for exploring emergent topological states and for advancing novel quantum device applications.

cond-mat.mes-hall↗

arXiv2025

Single-Frame Point-Pixel Registration via Supervised Cross-Modal Feature Matching

Point-pixel registration between LiDAR point clouds and camera images is a fundamental yet challenging task in autonomous driving and robotic perception. A key difficulty lies in the modality gap between unstructured point clouds and structured images, especially under sparse single-frame LiDAR settings. Existing methods typically extract features separately from point clouds and images, then rely on hand-crafted or learned matching strategies. This separate encoding fails to bridge the modality gap effectively, and more critically, these methods struggle with the sparsity and noise of single-frame LiDAR, often requiring point cloud accumulation or additional priors to improve reliability. Inspired by recent progress in detector-free matching paradigms, we revisit the projection-based approach and introduce the detector-free framework for direct point-pixel matching between LiDAR and camera views. To further enhance matching reliability, we introduce a repeatability scoring mechanism that acts as a soft visibility prior. This guides the network to suppress unreliable matches in regions with low intensity variation, improving robustness under sparse input. Extensive experiments on KITTI, nuScenes, and MIAS-LCEC-TF70 benchmarks demonstrate that our method achieves state-of-the-art performance, outperforming prior approaches on nuScenes (even those relying on accumulated point clouds), despite using only single-frame LiDAR.

cs.CV↗

arXiv2025

Exploration of Multi-Element Collaborative Research and Application for Modern Power System Based on Generative Large Models

The transition to intelligent, low-carbon power systems necessitates advanced optimization strategies for managing renewable energy integration, energy storage, and carbon emissions. Generative Large Models (GLMs) provide a data-driven approach to enhancing forecasting, scheduling, and market operations by processing multi-source data and capturing complex system dynamics. This paper explores the role of GLMs in optimizing load-side management, energy storage utilization, and electricity carbon, with a focus on Smart Wide-area Hybrid Energy Systems with Storage and Carbon (SGLSC). By leveraging spatiotemporal modeling and reinforcement learning, GLMs enable dynamic energy scheduling, improve grid stability, enhance carbon trading strategies, and strengthen resilience against extreme weather events. The proposed framework highlights the transformative potential of GLMs in achieving efficient, adaptive, and low-carbon power system operations.

eess.SY↗

arXiv2025

From Sparse to Dense: Camera Relocalization with Scene-Specific Detector from Feature Gaussian Splatting

This paper presents a novel camera relocalization method, STDLoc, which leverages Feature Gaussian as scene representation. STDLoc is a full relocalization pipeline that can achieve accurate relocalization without relying on any pose prior. Unlike previous coarse-to-fine localization methods that require image retrieval first and then feature matching, we propose a novel sparse-to-dense localization paradigm. Based on this scene representation, we introduce a novel matching-oriented Gaussian sampling strategy and a scene-specific detector to achieve efficient and robust initial pose estimation. Furthermore, based on the initial localization results, we align the query feature map to the Gaussian feature field by dense feature matching to enable accurate localization. The experiments on indoor and outdoor datasets show that STDLoc outperforms current state-of-the-art localization methods in terms of localization accuracy and recall.

cs.CV↗

arXiv2024

TrustUQA: A Trustful Framework for Unified Structured Data Question Answering

Natural language question answering (QA) over structured data sources such as tables and knowledge graphs have been widely investigated, especially with Large Language Models (LLMs) in recent years. The main solutions include question to formal query parsing and retrieval-based answer generation. However, current methods of the former often suffer from weak generalization, failing to dealing with multi-types of sources, while the later is limited in trustfulness. In this paper, we propose TrustUQA, a trustful QA framework that can simultaneously support multiple types of structured data in a unified way. To this end, it adopts an LLM-friendly and unified knowledge representation method called Condition Graph(CG), and uses an LLM and demonstration-based two-level method for CG querying. For enhancement, it is also equipped with dynamic demonstration retrieval. We have evaluated TrustUQA with 5 benchmarks covering 3 types of structured data. It outperforms 2 existing unified structured data QA methods. In comparison with the baselines that are specific to one data type, it achieves state-of-the-art on 2 of the datasets. Further more, we have demonstrated the potential of our method for more general QA tasks, QA over mixed structured data and QA across structured data. The code is available at https://github.com/zjukg/TrustUQA.

cs.CL↗

arXiv2024

VLM-Grounder: A VLM Agent for Zero-Shot 3D Visual Grounding

3D visual grounding is crucial for robots, requiring integration of natural language and 3D scene understanding. Traditional methods depending on supervised learning with 3D point clouds are limited by scarce datasets. Recently zero-shot methods leveraging LLMs have been proposed to address the data issue. While effective, these methods only use object-centric information, limiting their ability to handle complex queries. In this work, we present VLM-Grounder, a novel framework using vision-language models (VLMs) for zero-shot 3D visual grounding based solely on 2D images. VLM-Grounder dynamically stitches image sequences, employs a grounding and feedback scheme to find the target object, and uses a multi-view ensemble projection to accurately estimate 3D bounding boxes. Experiments on ScanRefer and Nr3D datasets show VLM-Grounder outperforms previous zero-shot methods, achieving 51.6% Acc@0.25 on ScanRefer and 48.0% Acc on Nr3D, without relying on 3D geometry or object priors. Codes are available at https://github.com/OpenRobotLab/VLM-Grounder .

cs.CV↗

arXiv2024

Rosette spectroscopic imaging for whole-brain metabolite mapping at 7T: acceleration potential and reproducibility

Whole-brain proton magnetic resonance spectroscopic imaging (1H-MRSI) is a non-invasive technique for assessing neurochemical distribution in the brain, offering valuable insights into brain functions and neural diseases. It greatly benefits from the improved SNR at ultrahigh field strengths ($\geq$7T). However, 1H-MRSI still faces several challenges, such as long acquisition time and severe signal contaminations from water and lipids. In this study, 2D and 3D short TR/TE 1H-FID-MRSI sequences using rosette trajectories were developed with spatial resolutions of 4.48$\times$4.48 mm$^2$ and 4.48$\times$4.48$\times$4.50 mm$^3$, respectively. Water signals were suppressed using an optimized Five-variable-Angle-gaussian-pulses-with-ShorT-total-duration of 76 ms (FAST) water suppression scheme, and lipid signals were removed using the L2 regularization method. Metabolic maps of major 1H metabolites were obtained within 5:40 min with 16 averages and 1 average for the 2D and 3D acquisitions, respectively. Excellent inter-session reproducibility was shown, with the coefficients of variance (CV) being lower than 6% for N-Acetyle-L-aspartic acid (NAA), Glutamate (Glu), Choline Chloride and glycerophosphocholine (tCho), Creatine and Phosphocreatine (tCr), and Glycine and Myo-inositol (Gly+Ins). To explore the potential of further accelerating the acquisition, compressed sensing was applied retrospectively to the 3D datasets. The structural similarity index (SSIM) remained above 0.85 and 0.8 until $R = 2$ and $R = 3$ for the metabolite maps of Glu, NAA, tCr, and tCho, indicating the possibility for further reduction of acquisition time to around 2min.

physics.med-ph↗

arXiv2024

A Toolbox for Optimization of Reconstruction and Post-processing Pipelines in In Vivo $^{31}$P MR Imaging

Background: X-nuclei imaging (e.g. 31P MRI) suffers low SNR due to lower gyromagnetic ratios and tissue concentrations than 1H MRI. Enhancing SNR requires high fields, advanced coils, optimized sequences, and sophisticated reconstruction methods. An optimized pipeline is essential for maximizing SNR in X-nuclei imaging. Purpose: To develop a toolbox of reconstruction and processing methods for fast 31P MR imaging, evaluating optimal pipelines for two data types: 31P-GRE and bSSFP-like MRF imaging. Methods: The toolbox includes coil combination (adaptive and whitened-SVD), k-space filtering, reconstruction (Kaiser-Bessel regridding with FFT and NUFFT), and denoising (compressed sensing and MP-PCA). 31P MR datasets were acquired at 7T using spiral encoding with a double-tuned birdcage coil and a 32-channel phased array. GRE parameters: readout 18.62ms, flip angle 59deg, TR 5400ms, TE 0.3ms. MRF used a bSSFP-like scheme: readout 9.12ms, TR 19.8ms, TE 0.3ms. Results: For GRE data, adaptive coil combination with Kaiser-Bessel regridding (AC-KB) achieved highest mean SNRs: 38 (phantom) and 3.4 (in vivo), outperforming others. For MRF data, whitened-SVD with Kaiser-Bessel regridding yielded an SNR of 18, surpassing AC-KB's 16. MP-PCA denoised GRE effectively; compressed sensing suited MRF data. Conclusion: We demonstrated a toolbox for optimizing fast 31P MR imaging pipelines. Optimal methods depend on acquisition and data. Kaiser-Bessel regridding outperforms NUFFT. Adaptive coil combination and MP-PCA denoising are preferred for GRE data; whitened-SVD and compressed sensing are superior for MRF data. Careful selection of processing methods is crucial, as conventional 1H MRI techniques may not suffice. The toolbox facilitates future investigation for other X-nuclei datasets.

physics.med-ph↗

arXiv2024

Reliable Academic Conference Question Answering: A Study Based on Large Language Model

As the development of academic conferences fosters global scholarly communication, researchers consistently need to obtain accurate and up-to-date information about academic conferences. Since the information is scattered, using an intelligent question-answering system to efficiently handle researchers' queries and ensure awareness of the latest advancements is necessary. Recently, Large Language Models (LLMs) have demonstrated impressive capabilities in question answering, and have been enhanced by retrieving external knowledge to deal with outdated knowledge. However, these methods fail to work due to the lack of the latest conference knowledge. To address this challenge, we develop the ConferenceQA dataset, consisting of seven diverse academic conferences. Specifically, for each conference, we first organize academic conference data in a tree-structured format through a semi-automated method. Then we annotate question-answer pairs and classify the pairs into four different types to better distinguish their difficulty. With the constructed dataset, we further propose a novel method STAR (STructure-Aware Retrieval) to improve the question-answering abilities of LLMs, leveraging inherent structural information during the retrieval process. Experimental results on the ConferenceQA dataset show the effectiveness of our retrieval method. The dataset and code are available at https://github.com/zjukg/ConferenceQA.

cs.CL↗

arXiv2024

Fast 3D 31P B1+ mapping with a weighted stack of spiral trajectory at 7 Tesla

Purpose: Phosphorus Magnetic Resonance Spectroscopy (31P MRS) enables non-invasive assessment of energy metabolism, yet its application is hindered by sensitivity limitations. To overcome this, often high magnetic fields are used, leading to challenges such as spatial B_1^+ inhomogeneity and therefore the need for accurate flip angle determination in accelerated acquisitions with short repetition times (T_R). In response to these challenges, we propose a novel short T_R and look-up table-based Double-Angle Method for fast 3D 31P B_1^+ mapping (fDAM). Methods: Our method incorporates 3D weighted stack of spiral gradient echo acquisitions and a frequency-selective pulse to enable efficient B_1^+ mapping based on the phosphocreatine signal at 7T. Protocols were optimised using simulations and validated through phantom experiments. The method was validated in phantom experiments and skeletal muscle applications using a birdcage 1H/31P volume coil. Results: The results of fDAM were compared to the classical DAM (cDAM). A good correlation (r=0.94) was obtained between the two B_1^+ maps. A 3D 31P B_1^+ mapping in the human calf muscle was achieved in about 10 min using a birdcage volume coil, with a 20% extended coverage relative to that of the cDAM (24 min). fDAM also enabled the first full brain coverage 31P 3D B_1^+ mapping in approx. 10 min using a 1 Tx/ 32 Rx coil. Conclusion: fDAM is an efficient method for 31P 3D B_1^+ mapping, showing promise for future applications in rapid 31P MRSI.

physics.med-ph↗

arXiv2024

Start from Zero: Triple Set Prediction for Automatic Knowledge Graph Completion

Knowledge graph (KG) completion aims to find out missing triples in a KG. Some tasks, such as link prediction and instance completion, have been proposed for KG completion. They are triple-level tasks with some elements in a missing triple given to predict the missing element of the triple. However, knowing some elements of the missing triple in advance is not always a realistic setting. In this paper, we propose a novel graph-level automatic KG completion task called Triple Set Prediction (TSP) which assumes none of the elements in the missing triples is given. TSP is to predict a set of missing triples given a set of known triples. To properly and accurately evaluate this new task, we propose 4 evaluation metrics including 3 classification metrics and 1 ranking metric, considering both the partial-open-world and the closed-world assumptions. Furthermore, to tackle the huge candidate triples for prediction, we propose a novel and efficient subgraph-based method GPHT that can predict the triple set fast. To fairly compare the TSP results, we also propose two types of methods RuleTensor-TSP and KGE-TSP applying the existing rule- and embedding-based methods for TSP as baselines. During experiments, we evaluate the proposed methods on two datasets extracted from Wikidata following the relation-similarity partial-open-world assumption proposed by us, and also create a complete family data set to evaluate TSP results following the closed-world assumption. Results prove that the methods can successfully generate a set of missing triples and achieve reasonable scores on the new task, and GPHT performs better than the baselines with significantly shorter prediction time. The datasets and code for experiments are available at https://github.com/zjukg/GPHT-for-TSP.

cs.AI↗

arXiv2024

Online,Target-Free LiDAR-Camera Extrinsic Calibration via Cross-Modal Mask Matching

LiDAR-camera extrinsic calibration (LCEC) is crucial for data fusion in intelligent vehicles. Offline, target-based approaches have long been the preferred choice in this field. However, they often demonstrate poor adaptability to real-world environments. This is largely because extrinsic parameters may change significantly due to moderate shocks or during extended operations in environments with vibrations. In contrast, online, target-free approaches provide greater adaptability yet typically lack robustness, primarily due to the challenges in cross-modal feature matching. Therefore, in this article, we unleash the full potential of large vision models (LVMs), which are emerging as a significant trend in the fields of computer vision and robotics, especially for embodied artificial intelligence, to achieve robust and accurate online, target-free LCEC across a variety of challenging scenarios. Our main contributions are threefold: we introduce a novel framework known as MIAS-LCEC, provide an open-source versatile calibration toolbox with an interactive visualization interface, and publish three real-world datasets captured from various indoor and outdoor environments. The cornerstone of our framework and toolbox is the cross-modal mask matching (C3M) algorithm, developed based on a state-of-the-art (SoTA) LVM and capable of generating sufficient and reliable matches. Extensive experiments conducted on these real-world datasets demonstrate the robustness of our approach and its superior performance compared to SoTA methods, particularly for the solid-state LiDARs with super-wide fields of view.

cs.RO↗