arXiv subjects

Jie Zhao

Publications and source records attributed to Jie Zhao.

At least 73 records · Page 4Linked to original sources

arXiv2025

Imitation Learning Policy based on Multi-Step Consistent Integration Shortcut Model

The wide application of flow-matching methods has greatly promoted the development of robot imitation learning. However, these methods all face the problem of high inference time. To address this issue, researchers have proposed distillation methods and consistency methods, but the performance of these methods still struggles to compete with that of the original diffusion models and flow-matching models. In this article, we propose a one-step shortcut method with multi-step integration for robot imitation learning. To balance the inference speed and performance, we extend the multi-step consistency loss on the basis of the shortcut model, split the one-step loss into multi-step losses, and improve the performance of one-step inference. Secondly, to solve the problem of unstable optimization of the multi-step loss and the original flow-matching loss, we propose an adaptive gradient allocation method to enhance the stability of the learning process. Finally, we evaluate the proposed method in two simulation benchmarks and five real-world environment tasks. The experimental results verify the effectiveness of the proposed algorithm.

cs.RO↗

arXiv2025

Stock Prediction via a Dual Relation Fusion Network incorporating Static and Dynamic Relations

Accurate modeling of inter-stock relationships is critical for stock price forecasting. However, existing methods predominantly focus on single-state relationships, neglecting the essential complementarity between dynamic and static inter-stock relations. To solve this problem, we propose a Dual Relation Fusion Network (DRFN) to capture the long-term relative stability of stock relation structures while retaining the flexibility to respond to sudden market shifts. Our approach features a novel relative static relation component that models time-varying long-term patterns and incorporates overnight informational influences. We capture dynamic inter-stock relationships through distance-aware mechanisms, while evolving long-term structures via recurrent fusion of dynamic relations from the prior day with the pre-defined static relations. Experiments demonstrate that our method significantly outperforms the baselines across different markets, with high sensitivity to the co-movement of relational strength and stock price.

cs.LG↗

arXiv2025

Multidomain Evolutionary Optimization on Combinatorial Problems in Complex Networks

Knowledge transfer-based evolutionary optimization has garnered significant attention, such as in multi-task evolutionary optimization (MTEO), which aims to solve complex problems by simultaneously optimizing multiple tasks. While this emerging paradigm has been primarily focusing on task similarity, there remains a hugely untapped potential in harnessing the shared characteristics between different domains. For example, real-world complex systems usually share the same characteristics, such as the power-law rule, small-world property and community structure, thus making it possible to transfer solutions optimized in one system to another to facilitate the optimization. Drawing inspiration from this observation of shared characteristics within complex systems, we present a novel framework, multi-domain evolutionary optimization (MDEO). First, we propose a community-level measurement of graph similarity to manage the knowledge transfer among domains. Furthermore, we develop a graph learning-based network alignment model that serves as the conduit for effectively transferring solutions between different domains. Moreover, we devise a self-adaptive mechanism to determine the number of transferred solutions from different domains, and introduce a knowledge-guided mutation mechanism that adaptively redefines mutation candidates to facilitate the utilization of knowledge from other domains. To evaluate its performance, we use a challenging combinatorial problem known as adversarial link perturbation as the primary illustrative optimization task. Experiments on multiple real-world networks of different domains demonstrate the superiority of the proposed framework in efficacy compared to classical evolutionary optimization.

cs.NE↗

arXiv2025

Can Large Language Models Be Trusted as Evolutionary Optimizers for Network-Structured Combinatorial Problems?

Large Language Models (LLMs) have shown strong capabilities in language understanding and reasoning across diverse domains. Recently, there has been increasing interest in utilizing LLMs not merely as assistants in optimization tasks, but as primary optimizers, particularly for network-structured combinatorial problems. However, before LLMs can be reliably deployed in this role, a fundamental question must be addressed: Can LLMs iteratively manipulate solutions that consistently adhere to problem constraints? In this work, we propose a systematic framework to evaluate the capability of LLMs to engage with problem structures. Rather than treating the model as a black-box generator, we adopt the commonly used evolutionary optimizer (EVO) and propose a comprehensive evaluation framework that rigorously assesses the output fidelity of LLM-based operators across different stages of the evolutionary process. To enhance robustness, we introduce a hybrid error-correction mechanism that mitigates uncertainty in LLMs outputs. Moreover, we explore a cost-efficient population-level optimization strategy that significantly improves efficiency compared to traditional individual-level approaches. Extensive experiments on a representative node-level combinatorial network optimization task demonstrate the effectiveness, adaptability, and inherent limitations of LLM-based EVO. Our findings present perspectives on integrating LLMs into evolutionary computation and discuss paths that may support scalable and context-aware optimization in networked systems.

cs.NE↗

arXiv2025

Incorporating Si into Sb2Se3: Tailoring Optical Phase Change Materials via Nanocomposites

Chalcogenide-based optical phase change materials (OPCMs) exhibit a large contrast in refractive index when reversibly switched between their stable amorphous and crystalline states. OPCMs have rapidly gained attention due to their versatility as nonvolatile amplitude or phase modulators in various photonic devices. However, open challenges remain, such as achieving reliable response and transparency spanning into the visible spectrum, a combination of properties in which current broadband OPCMs (e.g., Ge2Sb2Se4Te1, Sb2Se3, or Sb2S3) fall short. Discovering novel materials or engineering existing ones is, therefore, crucial in extending the application scope of OPCMs. Here, we use magnetron co-sputtering to study the effects of Si doping into Sb2Se3. We employ ellipsometry, X-ray diffraction, Raman spectroscopy, and scanning and transmission electron microscopy to investigate the effects of Si doping on the optical properties and crystal structure and compare these results with those from first principles calculations. Moreover, we study the crystallization and melt-quenching of thin films via nano-differential scanning calorimetry (NanoDSC). Our experiments demonstrate that 20% Si doping increases the transparency window in both states, specifically to 800 nm (1.55 eV) in the amorphous phase, while reducing power consumption by lowering the melting temperature. However, this reduction comes at the cost of reducing the refractive index contrast between states and slowing the kinetics of the phase transition.

cond-mat.mtrl-sci↗

arXiv2025

Design and Control of a Perching Drone Inspired by the Prey-Capturing Mechanism of Venus Flytrap

The endurance and energy efficiency of drones remain critical challenges in their design and operation. To extend mission duration, numerous studies explored perching mechanisms that enable drones to conserve energy by temporarily suspending flight. This paper presents a new perching drone that utilizes an active flexible perching mechanism inspired by the rapid predation mechanism of the Venus flytrap, achieving perching in less than 100 ms. The proposed system is designed for high-speed adaptability to the perching targets. The overall drone design is outlined, followed by the development and validation of the biomimetic perching structure. To enhance the system stability, a cascade extended high-gain observer (EHGO) based control method is developed, which can estimate and compensate for the external disturbance in real time. The experimental results demonstrate the adaptability of the perching structure and the superiority of the cascaded EHGO in resisting wind and perching disturbances.

cs.RO↗

arXiv2025

Visual Evolutionary Optimization on Graph-Structured Combinatorial Problems with MLLMs: A Case Study of Influence Maximization

Graph-structured combinatorial problems in complex networks are prevalent in many domains, and are computationally demanding due to their complexity and non-linear nature. Traditional evolutionary algorithms (EAs), while robust, often face obstacles due to content-shallow encoding limitations and lack of structural awareness, necessitating hand-crafted modifications for effective application. In this work, we introduce an original framework, visual evolutionary optimization (VEO), leveraging multimodal large language models (MLLMs) as the backbone evolutionary optimizer in this context. Specifically, we propose a context-aware encoding scheme, representing the solution of the network as an image. In this manner, we can utilize MLLMs' image processing capabilities to intuitively comprehend network configurations, thus enabling machines to solve these problems in a human-like way. We develop MLLM-based operators tailored for various evolutionary optimization stages, including initialization, crossover, and mutation. Furthermore, we propose that graph sparsification can effectively enhance the applicability and scalability of VEO on large-scale networks, owing to the scale-free nature of real-world networks. We demonstrate the effectiveness of our method using a well-known task in complex networks, influence maximization, and validate it on eight different real-world networks of various structures. The results confirm VEO's reliability and enhanced effectiveness compared to traditional evolutionary optimization.

cs.NE↗

arXiv2025

GDLLM: A Global Distance-aware Modeling Approach Based on Large Language Models for Event Temporal Relation Extraction

In Natural Language Processing(NLP), Event Temporal Relation Extraction (ETRE) is to recognize the temporal relations of two events. Prior studies have noted the importance of language models for ETRE. However, the restricted pre-trained knowledge of Small Language Models(SLMs) limits their capability to handle minority class relations in imbalanced classification datasets. For Large Language Models(LLMs), researchers adopt manually designed prompts or instructions, which may introduce extra noise, leading to interference with the model's judgment of the long-distance dependencies between events. To address these issues, we propose GDLLM, a Global Distance-aware modeling approach based on LLMs. We first present a distance-aware graph structure utilizing Graph Attention Network(GAT) to assist the LLMs in capturing long-distance dependency features. Additionally, we design a temporal feature learning paradigm based on soft inference to augment the identification of relations with a short-distance proximity band, which supplements the probabilistic information generated by LLMs into the multi-head attention mechanism. Since the global feature can be captured effectively, our framework substantially enhances the performance of minority relation classes and improves the overall learning ability. Experiments on two publicly available datasets, TB-Dense and MATRES, demonstrate that our approach achieves state-of-the-art (SOTA) performance.

cs.CL↗

arXiv2025

Sim-to-Real Transfer of Deep Reinforcement Learning Agents for Online Coverage Path Planning

Coverage path planning (CPP) is the problem of finding a path that covers the entire free space of a confined area, with applications ranging from robotic lawn mowing to search-and-rescue. While for known environments, offline methods can find provably complete paths, and in some cases optimal solutions, unknown environments need to be planned online during mapping. We investigate the suitability of continuous-space reinforcement learning (RL) for this challenging problem, and propose a computationally feasible egocentric map representation based on frontiers, as well as a novel reward term based on total variation to promote complete coverage. Compared to existing classical methods, this approach allows for a flexible path space, and enables the agent to adapt to specific environment characteristics. Meanwhile, the deployment of RL models on real robot systems is difficult. Training from scratch may be infeasible due to slow convergence times, while transferring from simulation to reality, i.e. sim-to-real transfer, is a key challenge in itself. We bridge the sim-to-real gap through a semi-virtual environment, including a real robot and real-time aspects, while utilizing a simulated sensor and obstacles to enable environment randomization and automated episode resetting. We investigate what level of fine-tuning is needed for adapting to a realistic setting. Through extensive experiments, we show that our approach surpasses the performance of both previous RL-based approaches and highly specialized methods across multiple CPP variations in simulation. Meanwhile, our method successfully transfers to a real robot. Our code implementation can be found online.

cs.RO↗

arXiv2025

Early Evolution of the Cavity and Core of a Coronal Mass Ejection in the Inner Corona

Coronal mass ejections (CMEs) typically exhibit a three-component structure in white-light (WL) coronagraphs. Utilizing the seamless observations of the inner corona ($\le$ 3 R$_\odot$), we have revealed the early evolution of the cavity and core of a CME starting at $\sim$18:20 UT on 2014 October 14. The CME originates from a hot channel (HC), which appears as the bright core and compresses the cavity in WL images. Specifically, most of the dark cavity in WL is filled by bright loop-like structures in 174 Å. The differential emission measure (DEM) analysis indicates that the electron temperature decreases from the core ($\sim$13.4 MK) to the cavity ($\sim$1.35 MK), and the CME cavity is significantly cooler than that enshrouding a prominence ($\ge$ 2 MK). The effective temperature of the cavity increases over time in general, probably due to the compression by the HC expansion. The evolution of the CME bright core includes slow-rise, fast-rise (up to $\sim$330 km s$^{-1}$), and residual-acceleration phases. The cavity exhibits an evolution similar to the core but lags by $\sim$4 minutes, with a lower speed peaking at $\sim$220 km s$^{-1}$. Moreover, the 2D radial speed distribution exhibits the highest speeds at the core apex. The kinematical results further confirm the compression of the cavity. The present event supports the new explanation of the CME structures, i.e., the magnetic flux rope (MFR), which is proxied by the HC, is only responsible for the core, while the cavity is likely a low-density region between the CME front and the MFR.

astro-ph.SR↗

arXiv2025

Progressive Bird's Eye View Perception for Safety-Critical Autonomous Driving: A Comprehensive Survey

Bird's-Eye-View (BEV) perception has become a foundational paradigm in autonomous driving, enabling unified spatial representations that support robust multi-sensor fusion and multi-agent collaboration. As autonomous vehicles transition from controlled environments to real-world deployment, ensuring the safety and reliability of BEV perception in complex scenarios - such as occlusions, adverse weather, and dynamic traffic - remains a critical challenge. This survey provides the first comprehensive review of BEV perception from a safety-critical perspective, systematically analyzing state-of-the-art frameworks and implementation strategies across three progressive stages: single-modality vehicle-side, multimodal vehicle-side, and multi-agent collaborative perception. Furthermore, we examine public datasets encompassing vehicle-side, roadside, and collaborative settings, evaluating their relevance to safety and robustness. We also identify key open-world challenges - including open-set recognition, large-scale unlabeled data, sensor degradation, and inter-agent communication latency - and outline future research directions, such as integration with end-to-end autonomous driving systems, embodied intelligence, and large language models.

cs.RO↗

arXiv2025

Counter-propagating Entangled Photon Pairs from a Monolayer

Non-phase-matched spontaneous parametric down-conversion (SPDC) in atomically thin materials provides new degrees of freedom and enhanced quantum information capacity compared to conventional phase-matched sources. These systems emerged as promising platforms for quantum computing, communication, and imaging, with the potential to support higher-order nonlinear processes. However, direct observation of photon-pair emission from a monolayer has remained experimentally challenging. In this work, we theoretically modeled SPDC emission across the full angular space from a monolayer GaSe film and experimentally validated the model through measurements of both co- and counter-propagating photon pairs. We demonstrated two-photon quantum correlations in the telecom C-band from the thinnest SPDC source reported to date. The spatially symmetric, broadband emission predicted by theory was confirmed experimentally. Furthermore, we observed high-fidelity Bell states in the counter-propagating configuration, marking the first realization of polarization-entangled photon pairs from a monolayer. Our results revealed the emission characteristics of SPDC in the deeply subwavelength, non-phase-matched regime, and introduced atomically thin, counterpropagating SPDC as a scalable and integrable platform for programmable quantum state generation, extendable via moiré superlattice engineering.

quant-ph↗

arXiv2025

DD-DeepONet: Domain decomposition and DeepONet for solving partial differential equations in three application scenarios

In certain practical engineering applications, there is an urgent need to perform repetitive solving of partial differential equations (PDEs) in a short period. This paper primarily considers three scenarios requiring extensive repetitive simulations. These three scenarios are categorized based on whether the geometry, boundary conditions(BCs), or parameters vary. We introduce the DD-DeepONet, a framework with strong scalability, whose core concept involves decomposing complex geometries into simple structures and vice versa. We primarily study complex geometries composed of rectangles and cuboids, which have numerous practical applications. Simultaneously, stretching transformations are applied to simple geometries to solve shape-dependent problems. This work solves several prototypical PDEs in three scenarios, including Laplace, Poission, N-S, and drift-diffusion equations, demonstrating DD-DeepONet's computational potential. Experimental results demonstrate that DD-DeepONet reduces training difficulty, requires smaller datasets andVRAMper network, and accelerates solution acquisition.

math.NA↗

arXiv2025

Enhanced continuous-variable quantum key distribution protocol via adaptive signal processing

Quantum key distribution (QKD) provides a promising approach to secure communications, with continuous-variable QKD (CV-QKD) offering compatibility with existing telecommunication infrastructure. Despite this advantage, CV-QKD is limited by challenges such as losses in terrestrial fibres and atmospheric scintillation in free-space channels. We introduce a QKD protocol that surpasses the optimal Gaussian modulated CV-QKD (GG02) protocol by utilising probabilistic filters without known physical representation. Our approach employs a Gaussian filter at Alice's station and a non-Gaussian notch-like filter at Bob's station. Alice's filter optimises modulation variance to achieve key rates near the optimal GG02 performance, while Bob's filter adapts the effective channel conditions, which can result in higher key rates than the optimal GG02 protocol. Our security analysis avoids Gaussian extremality, accurately bounding Eve's information. The protocol dynamically optimises the secret-key rate for rapidly changing channels, such as terrestrial links and satellite-to-ground communications, and can extract keys in regions deemed non-secure by parameter estimation. Implemented at software level, our protocol requires no hardware modifications and can be integrated into existing QKD systems. Experimental results show a threefold increase in key rates over the optimal GG02 protocol, while simulations for Low Earth Orbit satellite quantum communications indicate a 400-fold increase compared to the non-optimised counterpart.

quant-ph↗

arXiv2025

DOFA-CLIP: Multimodal Vision-Language Foundation Models for Earth Observation

Earth observation (EO) spans a broad spectrum of modalities, including optical, radar, multispectral, and hyperspectral data, each capturing distinct environmental signals. However, current vision-language models in EO, particularly CLIP-based variants, remain confined to individual modalities, limiting generalization and scalability across diverse tasks. We present DOFA-CLIP (Dynamic-One-For-All CLIP), a unified vision-language foundation model that dynamically adapts to EO modalities with flexible spectral configurations through a single Transformer backbone. Our approach introduces three key contributions: 1) the construction of GeoLangBind-2M, a large-scale EO image-text dataset covering six heterogeneous modalities with rich natural language descriptions; 2) a novel training strategy called VECT (Vision-models Enhanced Contrastive Text-image pretraining), which enhances the spatial awareness of CLIP features with multiple vision foundation models; and 3) a Modality-aware Knowledge Agglomeration (MaKA) module that refines feature distillation with modality-specific awareness. DOFA-CLIP achieves state-of-the-art zero-shot performance across a wide range of EO benchmarks, including unseen modalities and a diverse number of input spectral bands. Together, these contributions establish a scalable foundation for multimodal EO understanding and open new avenues for integrating heterogeneous EO data with large language models. Code and datasets will be released. Code and datasets are publicly available.

cs.CV↗

arXiv2025

Robust Multimodal Large Language Models Against Modality Conflict

Despite the impressive capabilities of multimodal large language models (MLLMs) in vision-language tasks, they are prone to hallucinations in real-world scenarios. This paper investigates the hallucination phenomenon in MLLMs from the perspective of modality conflict. Unlike existing works focusing on the conflicts between model responses and inputs, we study the inherent conflicts in inputs from different modalities that place MLLMs in a dilemma and directly lead to hallucinations. We formally define the modality conflict and construct a dataset named Multimodal Modality Conflict (MMMC) to simulate this phenomenon in vision-language tasks. Three methods based on prompt engineering, supervised fine-tuning, and reinforcement learning are proposed to alleviate the hallucination caused by modality conflict. Extensive experiments are conducted on the MMMC dataset to analyze the merits and demerits of these methods. Our results show that the reinforcement learning method achieves the best performance in mitigating the hallucination under modality conflict, while the supervised fine-tuning method shows promising and stable performance. Our work sheds light on the unnoticed modality conflict that leads to hallucinations and provides more insights into the robustness of MLLMs.

cs.CV↗

arXiv2025

Exploiting Lightweight Hierarchical ViT and Dynamic Framework for Efficient Visual Tracking

Transformer-based visual trackers have demonstrated significant advancements due to their powerful modeling capabilities. However, their practicality is limited on resource-constrained devices because of their slow processing speeds. To address this challenge, we present HiT, a novel family of efficient tracking models that achieve high performance while maintaining fast operation across various devices. The core innovation of HiT lies in its Bridge Module, which connects lightweight transformers to the tracking framework, enhancing feature representation quality. Additionally, we introduce a dual-image position encoding approach to effectively encode spatial information. HiT achieves an impressive speed of 61 frames per second (fps) on the NVIDIA Jetson AGX platform, alongside a competitive AUC of 64.6% on the LaSOT benchmark, outperforming all previous efficient trackers.Building on HiT, we propose DyHiT, an efficient dynamic tracker that flexibly adapts to scene complexity by selecting routes with varying computational requirements. DyHiT uses search area features extracted by the backbone network and inputs them into an efficient dynamic router to classify tracking scenarios. Based on the classification, DyHiT applies a divide-and-conquer strategy, selecting appropriate routes to achieve a superior trade-off between accuracy and speed. The fastest version of DyHiT achieves 111 fps on NVIDIA Jetson AGX while maintaining an AUC of 62.4% on LaSOT.Furthermore, we introduce a training-free acceleration method based on the dynamic routing architecture of DyHiT. This method significantly improves the execution speed of various high-performance trackers without sacrificing accuracy. For instance, our acceleration method enables the state-of-the-art tracker SeqTrack-B256 to achieve a 2.68 times speedup on an NVIDIA GeForce RTX 2080 Ti GPU while maintaining the same AUC of 69.9% on the LaSOT.

cs.CV↗

arXiv2025

Fabrication of airbridges with gradient exposure

In superconducting quantum circuits, airbridges are critical for eliminating parasitic slotline modes of coplanar waveguide circuits and reducing crosstalks between direct current magnetic flux biases. Here, we present a technique for fabricating superconducting airbridges. With this technique, a single layer of photoresist is employed, and the gradient exposure process is used to define the profile of airbridges. In order to properly obtain the bridge profile, we design exposure dosage based on residual photoresist thickness and laser power calibrations. Compared with other airbridge fabrication techniques, the gradient exposure fabrication technique provides the ability to produce lossless superconducting airbridges with flexible size and, thus, is more suitable for large-scale superconducting quantum circuits. Furthermore, this method reduces the complexity of the fabrication process and provides a high fabrication yield.

quant-ph↗