arXiv subjects

Zhixuan Shen

Publications and source records attributed to Zhixuan Shen.

11 recordsLinked to original sources

arXiv2026

Plan in Sandbox, Navigate in Open Worlds: Learning Physics-Grounded Abstracted Experience for Embodied Navigation

Vision-Language Models (VLMs) have demonstrated exceptional general reasoning capabilities. However, their performance in embodied navigation remains hindered by a scarcity of aligned open-world vision and robot control data. Despite simulators providing a cost-effective alternative for data collection, the inherent reliance on photorealistic simulations often limits the transferability of learned policies. To this end, we propose \textit{\textbf{S}andbox-\textbf{A}bstracted \textbf{G}rounded \textbf{E}xperience} (\textbf{\textit{SAGE}}), a framework that enables agents to learn within a physics-grounded semantic abstraction rather than a photorealistic simulation, mimicking the human capacity for mental simulation where plans are rehearsed in simplified physics abstractions before execution. \textit{SAGE} system operates via three synergistic phases: (1) \textit{Genesis}: constructing diverse, physics-constrained semantic environments to bootstrap experience; (2) \textit{Evolution}: distilling experiences through Reinforcement Learning (RL), utilizing a novel asymmetric adaptive clipping mechanism to stabilize updates; (3) \textit{Navigation}: bridging the abstract policy to open-world control. We demonstrate that \textit{SAGE} significantly improves planner-assisted embodied navigation, achieving a 53.21\% LLM-Match Success Rate on A-EQA (+9.7\% over baseline), while showing encouraging transfer to physical indoor robot deployment.

cs.RO↗

arXiv2026

Plug-and-Play Label Map Diffusion for Universal Goal-Oriented Navigation

In embodied vision, Goal-Oriented Navigation (GON) requires robots to locate a specific goal within an unexplored environment. The primary challenge of GON arises from the need to construct a Bird's-Eye-View (BEV) map to understand the environment while simultaneously localizing an unobserved goal. Existing map-based methods typically employ self-centered semantic maps, often facing challenges such as reliance on complete maps or inconsistent semantic association. To this end, we propose Plug-and-Play Label Map Diffusion (PLMD), which defines a novel map completion diffusion model based on Denoising Diffusion Probabilistic Models (DDPM). PLMD generates obstacle and semantic labels for unobserved regions through a diffusion-based completion process, thereby enabling goal localization even in partially observed environments. Moreover, it mitigates inconsistent semantic association by leveraging structural consistency between known and unknown obstacle layouts and integrating obstacle priors into the semantic denoising process. By substituting predicted labels for unobserved regions, robots can accurately localize the specified objects. Extensive experiments demonstrate that PLMD \textbf{(I)} effectively expands the region of unknown maps, \textbf{(II)} integrates seamlessly into existing navigation strategies that rely on semantic maps, \textbf{(III)} achieves state-of-the-art performance on three GON tasks.

cs.RO↗

arXiv2025

Enhancing Multi-Robot Semantic Navigation Through Multimodal Chain-of-Thought Score Collaboration

Understanding how humans cooperatively utilize semantic knowledge to explore unfamiliar environments and decide on navigation directions is critical for house service multi-robot systems. Previous methods primarily focused on single-robot centralized planning strategies, which severely limited exploration efficiency. Recent research has considered decentralized planning strategies for multiple robots, assigning separate planning models to each robot, but these approaches often overlook communication costs. In this work, we propose Multimodal Chain-of-Thought Co-Navigation (MCoCoNav), a modular approach that utilizes multimodal Chain-of-Thought to plan collaborative semantic navigation for multiple robots. MCoCoNav combines visual perception with Vision Language Models (VLMs) to evaluate exploration value through probabilistic scoring, thus reducing time costs and achieving stable outputs. Additionally, a global semantic map is used as a communication bridge, minimizing communication overhead while integrating observational results. Guided by scores that reflect exploration trends, robots utilize this map to assess whether to explore new frontier points or revisit history nodes. Experiments on HM3D_v0.2 and MP3D demonstrate the effectiveness of our approach. Our code is available at https://github.com/FrankZxShen/MCoCoNav.git.

cs.RO↗

arXiv2024

GatedUniPose: A Novel Approach for Pose Estimation Combining UniRepLKNet and Gated Convolution

Pose estimation is a crucial task in computer vision, with wide applications in autonomous driving, human motion capture, and virtual reality. However, existing methods still face challenges in achieving high accuracy, particularly in complex scenes. This paper proposes a novel pose estimation method, GatedUniPose, which combines UniRepLKNet and Gated Convolution and introduces the GLACE module for embedding. Additionally, we enhance the feature map concatenation method in the head layer by using DySample upsampling. Compared to existing methods, GatedUniPose excels in handling complex scenes and occlusion challenges. Experimental results on the COCO, MPII, and CrowdPose datasets demonstrate that GatedUniPose achieves significant performance improvements with a relatively small number of parameters, yielding better or comparable results to models with similar or larger parameter sizes.

cs.CV↗

arXiv2024

GateAttentionPose: Enhancing Pose Estimation with Agent Attention and Improved Gated Convolutions

This paper introduces GateAttentionPose, an innovative approach that enhances the UniRepLKNet architecture for pose estimation tasks. We present two key contributions: the Agent Attention module and the Gate-Enhanced Feedforward Block (GEFB). The Agent Attention module replaces large kernel convolutions, significantly improving computational efficiency while preserving global context modeling. The GEFB augments feature extraction and processing capabilities, particularly in complex scenes. Extensive evaluations on COCO and MPII datasets demonstrate that GateAttentionPose outperforms existing state-of-the-art methods, including the original UniRepLKNet, achieving superior or comparable results with improved efficiency. Our approach offers a robust solution for pose estimation across diverse applications, including autonomous driving, human motion capture, and virtual reality.

cs.CV↗

arXiv2024

Adversarial Training with OCR Modality Perturbation for Scene-Text Visual Question Answering

Scene-Text Visual Question Answering (ST-VQA) aims to understand scene text in images and answer questions related to the text content. Most existing methods heavily rely on the accuracy of Optical Character Recognition (OCR) systems, and aggressive fine-tuning based on limited spatial location information and erroneous OCR text information often leads to inevitable overfitting. In this paper, we propose a multimodal adversarial training architecture with spatial awareness capabilities. Specifically, we introduce an Adversarial OCR Enhancement (AOE) module, which leverages adversarial training in the embedding space of OCR modality to enhance fault-tolerant representation of OCR texts, thereby reducing noise caused by OCR errors. Simultaneously, We add a Spatial-Aware Self-Attention (SASA) mechanism to help the model better capture the spatial relationships among OCR tokens. Various experiments demonstrate that our method achieves significant performance improvements on both the ST-VQA and TextVQA datasets and provides a novel paradigm for multimodal adversarial training.

cs.CV↗

arXiv2017

Defects controlled hole doping and multi-valley transport in SnSe single crystals

SnSe is a promising thermoelectric material with record-breaking figure of merit, \textit{i.e., ZT}. As a semiconductor, optimal electrical dosage is the key challenge to maximize \textit{ZT} in SnSe. However, to date a comprehensive understanding of the electronic structure and most critically, the self-hole doping mechanism in SnSe is still absent. Here, we report the highly anisotropic electronic structure of SnSe investigated by both angle-resolved photoemission spectroscopy and quantum transport, in which a unique "\textit{pudding-mold}" shaped valence band with quasi-linear energy dispersion is revealed. We prove that the electrical doping in SnSe is extrinsically controlled by the formation of SnSe$_{2}$ micro-domains induced by local phase segregation. Using different growth methods and conditions, we have achieved wide tuning of hole doping in SnSe, ranging from intrinsic semiconducting behaviour to typical metal with carrier density of $1.23\times 10^{18}$ cm$^{-3}$ at room temperature. The resulting multi-valley transport in $p$-SnSe is characterized by non-saturating weak localization along the armchair axis, due to strong intervalley scattering enhanced by in-plane ferroelectric dipole field of the puckering lattice. Strikingly, quantum oscillations of magnetoresistance reveal three-dimensional electronic structure with unusual interlayer coupling strength in $p$-SnSe, which is correlated to the interweaving of SnSe individual layers by unique point dislocation defects. Our results suggest that defect engineering may provide versatile routes in improving the thermoelectric performance of the SnSe family.

cond-mat.mtrl-sci↗

arXiv2016

Negative Magnetoresistance in Weyl semimetals NbAs and NbP: Intrinsic Chiral Anomaly and Extrinsic Effects

Chiral anomaly induced negative magnetoresistance (NMR) has been widely used as a critical transport evidence on the existence of Weyl fermions in topological semimetals. In this mini review, we discuss the general observation of the NMR phenomena in non-centrosymmetric NbP and NbAs. We show that NMR can be contributed by intrinsic chiral anomaly of Weyl fermions and/or extrinsic effects, such as superimposition of Hall signals, field-dependent inhomogeneous current flow in the bulk, i.e. current jetting, and weak localization (WL) of coexistent trivial carriers. Such WL controlled NMR is heavily dependent on sample quality, and is characterized by pronounced crossover from positive to negative MR growth at elevated temperatures, as a result of the competition between the phase coherence time and the spin-orbital scattering constant of the bulk trivial pockets. Thus, the correlation of NMR and chiral anomaly needs to be scrutinized, without the support of other complimentary techniques. Due to the lifting of spin degeneracy, the spin orientations of Weyl fermions are either parallel or antiparallel to the momentum, a unique physical property known as helicity. The conservation of helicity provides strong protection for the transport of Weyl fermions, which can only be effectively scattered by magnetic impurities. Chemical doping of magnetic and non-magnetic impurities are thus more convincing in probing the existence of Weyl fermions than the NMR method.

cond-mat.mtrl-sci↗

arXiv2016

Negative Magnetoresistance in Topological Semimetals of Transition-Metal Dipnictides with Nontrivial Z2 Indices

Negative magnetoresistance (NMR) induced by the Adler-Bell-Jackiw anomaly is regarded as the most prominent quantum signature of Weyl semimetals when electrical field $E$ is collinear with the external magnetic field $B$. In this article, we report universal NMR in nonmagnetic, centrosymmetric transition metal dipnictides MPn$_{2}$ (M=Nb and Ta; Pn=As and Sb), in which the existence of Weyl fermions can be explicitly excluded. Using temperature-dependent magnetoresistance, Hall and thermoelectric coefficients of Nernst and Seebeck effects, we determine that the emergence of the NMR phenomena in MPn$_{2}$ is coincident with a Lifshitz transition, corresponding to the formation of unique electron-hole-electron ($e$-$h$-$e$) pockets along the $I-L-I'$ direction. First-principles calculations reveal that, along the $I-L-I'$ line, the $d_{xy}$ and $d_{x^{2}-y^{2}}$ orbitals of the transition metal form tilted nodal rings of band crossing well below the Fermi level. Strong spin-orbital coupling gaps all the crossing points and creates the characteristic $e$-$h$-$e$ structure, making MPn$_{2}$ a topological semimetal with $\mathbb{Z}_2$ indices of [0;(111)]. By excluding the weak localization contribution of the bulk states, we conclude that the universal NMR in MPn$_{2}$ may have an exotic origin in topological surface states, which appears in pairs with opposite spin-momentum locking on nontrivial surfaces.

cond-mat.str-el↗

arXiv2016

Helicity protected ultrahigh mobility Weyl fermions in NbP

Non-centrosymmetric transition metal monopnictides, including TaAs, TaP, NbAs, and NbP, are emergent topological Weyl semimetals (WSMs) hosting exotic relativistic Weyl fermions. In this letter, we elucidate the physical origin of the unprecedented charge carrier mobility of NbP, which can reach $1\times10^{7}$ cm $^{2}$V$^{-1}$s$^{-1}$ at 1.5 K. Angle- and temperature-dependent quantum oscillations, supported by density function theory calculations, reveal that NbP has the coexistence of p- and n-type WSM pockets in the $k_{z}$=1.16$π$/c plane (W1-WSM) and in the $k_{z}$=0 plane near the high symmetry points $Σ$ (W2-WSM), respectively. Uniquely, each W2-WSM pocket forms a large dumbbell-shaped Fermi surface (FS) enclosing two neighboring Weyl nodes with the opposite chirality. The magneto-transport in NbP is dominated by these highly anisotropic W2-WSM pockets, in which Weyl fermions are well protected from defect backscattering by real spin conservation associated to the chiral nodes. However, with a minimal doping of $\sim$1\% Cr, the mobility of NbP is degraded by more than two order of magnitude, due to the invalid of helicity protection to magnetic impurities. Helicity protected Weyl fermion transport is also manifested in chiral anomaly induced negative magnetoresistance, controlled by the W1-WSM states. In the quantum regime below 10 K, the intervalley scattering time by impurities becomes a large constant, producing the sharp and nearly identical conductivity enhancement at low magnetic field.

cond-mat.mes-hall↗

arXiv2016

Topological phase transition induced extreme magnetoresistance in TaSb$_{2}$

We report extremely large positive magnetoresistance of 1.72 million percent in single crystal TaSb$_{2}$ at moderate conditions of 1.5 K and 15 T. The quadratic growth of magnetoresistance (MR $\propto\,B^{1.96}$) is not saturating up to 15 T, a manifestation of nearly perfect compensation with $<0.1\%$ mismatch between electron and hole pockets in this semimetal. The compensation mechanism is confirmed by temperature-dependent MR, Hall and thermoelectric coefficients of Nernst and Seebeck, revealing two pronounced Fermi surface reconstruction processes without spontaneous symmetry breaking, \textit{i.e.} Lifshitz transitions, at around 20 K and 60 K, respectively. Using quantum oscillations of magnetoresistance and magnetic susceptibility, supported by density-functional theory calculations, we determined that the main hole Fermi surface of TaSb$_{2}$ forms a unique shoulder structure along the $F-L$ line. The flat band top of this shoulder pocket is just a few meV above the Fermi level, leading to the observed topological phase transition at 20 K when the shoulder pocket disappears. Further increase in temperature pushes the Fermi level to the band top of the main hole pocket, induced the second Lifshitz transition at 60 K when hole pocket vanishes completely.

cond-mat.str-el↗