arXiv subjects

Jie Feng

Publications and source records attributed to Jie Feng.

At least 55 records · Page 3Linked to original sources

arXiv2026

Resonance4D: Frequency-Domain Motion Supervision for Preset-Free Physical Parameter Learning in 4D Dynamic Physical Scene Simulation

Physics-driven 4D dynamic simulation from static 3D scenes remains constrained by an overlooked contradiction: reliable motion supervision often relies on online video diffusion or optical-flow pipelines whose computational cost exceeds that of the simulator itself. Existing methods further simplify inverse physical modeling by optimizing only partial material parameters, limiting realism in scenes with complex materials and dynamics. We present Resonance4D, a physics-driven 4D dynamic simulation framework that couples 3D Gaussian Splatting with the Material Point Method through lightweight yet physically expressive supervision. Our key insight is that dynamic consistency can be enforced without dense temporal generation by jointly constraining motion in complementary domains. To this end, we introduce Dual-domain Motion Supervision (DMS), which combines spatial structural consistency for local deformation with frequency-domain spectral consistency for oscillatory and global dynamic patterns, substantially reducing training cost and memory overhead while preserving physically meaningful motion cues. To enable stable full-parameter physical recovery, we further combine zero-shot text-prompted segmentation with simulation-guided initialization to automatically decompose Gaussians into object-part-level regions and support joint optimization of full material parameters. Experiments on both synthetic and real scenes show that Resonance4D achieves strong physical fidelity and motion consistency while reducing peak GPU memory from over 35\,GB to around 20\,GB, enabling high-fidelity physics-driven 4D simulation on a single consumer-grade GPU.

cs.CV↗

arXiv2026

Decouple and Rectify: Semantics-Preserving Structural Enhancement for Open-Vocabulary Remote Sensing Segmentation

Open-vocabulary semantic segmentation in the remote sensing (RS) field requires both language-aligned recognition and fine-grained spatial delineation. Although CLIP offers robust semantic generalization, its global-aligned visual representations inherently struggle to capture structural details. Recent methods attempt to compensate for this by introducing RS-pretrained DINO features. However, these methods treat CLIP representations as a monolithic semantic space and cannot localize where structural enhancement is required, failing to effectively delineate boundaries while risking the disruption of CLIP's semantic integrity. To address this limitation, we propose DR-Seg, a novel decouple-and-rectify framework in this paper. Our method is motivated by the key observation that CLIP feature channels exhibit distinct functional heterogeneity rather than forming a uniform semantic space. Building on this insight, DR-Seg decouples CLIP features into semantics-dominated and structure-dominated subspaces, enabling targeted structural enhancement by DINO without distorting language-aligned semantics. Subsequently, a prior-driven graph rectification module injects high-fidelity structural priors under DINO guidance to form a refined branch, while an uncertainty-guided adaptive fusion module dynamically integrates this refined branch with the original CLIP branch for final prediction. Comprehensive experiments across eight benchmarks demonstrate that DR-Seg establishes a new state-of-the-art.

cs.CV↗

arXiv2026

Are VLMs Lost Between Sky and Space? LinkS$^2$Bench for UAV-Satellite Dynamic Cross-View Spatial Intelligence

Synergistic spatial intelligence between UAVs and satellites is indispensable for emergency response and security operations, as it uniquely integrates macro-scale global coverage with dynamic, real-time local perception. However, the capacity of Vision-Language Models (VLMs) to master this complex interplay remains largely unexplored. This gap persists primarily because existing benchmarks are confined to isolated Unmanned Aerial Vehicle (UAV) videos or static satellite imagery, failing to evaluate the dynamic local-to-global spatial mapping essential for comprehensive cross-view reasoning. To bridge this gap, we introduce LinkS$^2$Bench, the first comprehensive benchmark designed to evaluate VLMs' wide-area, dynamic cross-view spatial intelligence. LinkS$^2$Bench links 1,022 minutes of dynamic UAV footage with high-resolution satellite imagery covering over 200 km$^2$. Through an LMM-assisted pipeline and rigorous human annotation, we constructed 17.9k high-quality question-answer pairs comprising 12 fine-grained tasks across four dimensions: perception, localization, relation, and reasoning. Evaluations of 18 representative VLMs reveal a substantial gap compared to human baselines, identifying accurate cross-view dynamic alignment as the critical bottleneck. To alleviate this, we design a Cross-View Alignment Adapter, demonstrating that explicit alignment significantly improves model performance. Furthermore, fine-tuning experiments underscore the potential of LinkS$^2$Bench in advancing VLM adaptation for complex spatial reasoning.

cs.CV↗

arXiv2026

Error Growth Dynamic and Predictability of Tropical Cyclone in Machine Learning Weather Prediction Model

Predictability analysis, which focuses on perturbation growth dynamic, is a key problem in both weather and climate prediction. Among all perturbations, the conditional nonlinear optimal perturbation (CNOP) leads to maximum uncertainties in forecasts, which is fundamentally important for theoretical studies and applications. Traditionally, CNOPs are solved through iterative optimization of numerical weather prediction (NWP) systems. Their large computational demands pose significant challenges to long-term predictability analysis. In our study, using a fast and accurate Artificial intelligence (AI) model, i.e. FuXi, a low-cost optimization framework for solving 5-day tropical cyclone (TC) CNOP is developed. For the first time, CNOPs that achieve the optimal (i.e., fastest) nonlinear development of long-term TC forecast errors are solved, with their optimality and physical explainability verified. Results demonstrate that perturbations with specific spatial structures undergo significant development. In both AI and NWP models, AI-based CNOPs exhibit rapid and physically consistent error growth across diverse TC cases, faster compared to random and lagged forecast perturbations. Furthermore, sensitivity analysis reveals that far-environment systems and processes are more crucial for long-term TC forecasts. Structural analyses of the CNOP emphasizes the interactions between TC internal and external processes for rapid perturbation growth. The success derivation of AI-based CNOPs, with their rapid growth and physical explainability verified in both AI and NWP models, suggests that AI models can capture the most rapidly growing perturbation patterns and their subsequent nonlinear evolution. Thus, potential of AI models is highlighted for advancing atmospheric predictability researches, including theoretical analysis, targeting observations and ensemble forecasts.

physics.ao-ph↗

arXiv2026

Experimental Determination of Gamma-Ray Polarization in Strong-Field Nonlinear Compton Scattering

The polarization of gamma rays produced in strong-field quantum electrodynamics (SFQED) is a fundamental and long-standing prediction, the verification of which has remained elusive, limiting both foundational tests and applications. Here, we report the first experimental measurement of gamma-ray polarization generated via all-optical nonlinear Compton scattering. Colliding a laser-wakefield-accelerated electron beam with an intense counter-propagating laser pulse reflected from a plasma mirror, we produce bright gamma rays in the strong-field regime ($a_0 > 1$). For gamma rays with $a_0 \approx 3$, a linear polarization degree of $\sim 50\%$ is measured via the azimuthal asymmetry of photoneutrons from a deuterium target, and independently verified by a Compton polarimeter.The results show excellent agreement with SFQED calculations employing the locally monochromatic approximation, while diverging from predictions based on the locally constant field approximation, highlighting the importance of quantum interference effects in this regime. Our work provides experimental evidence for polarization dynamics in SFQED, supports a key prediction of nonperturbative QED, and paves the way for compact, laser-driven sources of polarized gamma rays.}

physics.plasm-ph↗

arXiv2026

CityLens: Evaluating Large Vision-Language Models for Urban Socioeconomic Sensing

Understanding urban socioeconomic conditions through visual data is a challenging yet essential task for sustainable urban development and policy planning. In this work, we introduce \textit{CityLens}, a comprehensive benchmark designed to evaluate the capabilities of Large Vision-Language Models (LVLMs) in predicting socioeconomic indicators from satellite and street view imagery. We construct a multi-modal dataset covering a total of 17 globally distributed cities, spanning 6 key domains: economy, education, crime, transport, health, and environment, reflecting the multifaceted nature of urban life. Based on this dataset, we define 11 prediction tasks and utilize 3 evaluation paradigms: Direct Metric Prediction, Normalized Metric Estimation, and Feature-Based Regression. We benchmark 17 state-of-the-art LVLMs across these tasks. These make CityLens the most extensive socioeconomic benchmark to date in terms of geographic coverage, indicator diversity, and model scale. Our results reveal that while LVLMs demonstrate promising perceptual and reasoning capabilities, they still exhibit limitations in predicting urban socioeconomic indicators. CityLens provides a unified framework for diagnosing these limitations and guiding future efforts in using LVLMs to understand and predict urban socioeconomic patterns. The code and data are available at https://github.com/tsinghua-fib-lab/CityLens.

cs.AI↗

arXiv2026

Explore Intrinsic Geometry for Query-based Tiny and Oriented Object Detector with Momentum-based Bipartite Matching

Recent query-based detectors have achieved remarkable progress, yet their performance remains constrained when handling objects with arbitrary orientations, especially for tiny objects capturing limited texture information. This limitation primarily stems from the underutilization of intrinsic geometry during pixel-based feature decoding and the occurrence of inter-stage matching inconsistency caused by stage-wise bipartite matching. To tackle these challenges, we present IGOFormer, a novel query-based oriented object detector that explicitly integrates intrinsic geometry into feature decoding and enhances inter-stage matching stability. Specifically, we design an Intrinsic Geometry-aware Decoder, which enhances the object-related features conditioned on an object query by injecting complementary geometric embeddings extrapolated from their correlations to capture the geometric layout of the object, thereby offering a critical geometric insight into its orientation. Meanwhile, a Momentum-based Bipartite Matching scheme is developed to adaptively aggregate historical matching costs by formulating an exponential moving average with query-specific smoothing factors, effectively preventing conflicting supervisory signals arising from inter-stage matching inconsistency. Extensive experiments and ablation studies demonstrate the superiority of our IGOFormer for aerial oriented object detection, achieving an AP$_{50}$ score of 78.00\% on DOTA-V1.0 using Swin-T backbone under the single-scale setting. The code will be made publicly available.

cs.CV↗

arXiv2026

Efficient Policy Adaptation for Voltage Control Under Unknown Topology Changes

Reinforcement learning (RL) has shown great potential for designing voltage control policies, but their performance often degrades under changing system conditions such as topology reconfigurations and load variations. We introduce a topology-aware online policy optimization framework that leverages data-driven estimation of voltage-reactive power sensitivities to achieve efficient policy adaptation. Exploiting the sparsity of topology-switching events, where only a few lines change at a time, our method efficiently detects topology changes and identifies the affected lines and parameters, enabling fast and accurate sensitivity updates without recomputing the full sensitivity matrix. The estimated sensitivity is subsequently used for online policy optimization of a pre-trained neural-network-based RL controller. Simulations on both the IEEE 13-bus and SCE 56-bus systems demonstrate over 90 percent line identification accuracy, using only 15 data points. The proposed method also significantly improves voltage regulation performance compared with non-adaptive policies and adaptive policies that rely on regression-based online optimization methods for sensitivity estimation.

eess.SY↗

arXiv2026

CTRL-RAG: Contrastive Likelihood Reward Based Reinforcement Learning for Context-Faithful RAG Models

With the growing use of Retrieval-Augmented Generation (RAG), training large language models (LLMs) for context-sensitive reasoning and faithfulness is increasingly important. Existing RAG-oriented reinforcement learning (RL) methods rely on external rewards that often fail to evaluate document faithfulness, and may misjudge similar answers in open-domain settings. In addition, there is no RAG-based selfreward mechanism. Moreover, although such a mechanism could in principle estimate answer confidence given documents, the absence of objective feedback in a self-judgment can cause hallucination accumulation and eventual model collapse. To tackle these issues, we propose a novel "internal-external" hybrid reward framework centered on a Contrastive Likelihood Reward (CLR). CLR directly optimizes the log-likelihood gap between responses conditioned on prompts with and without supporting evidence. This encourages the model to extract relevant evidence and increases its confidence when grounded in a specific context. Experiments show that our method (used alone or combined with external correctness rewards) achieves strong performance on singlehop, multi-hop, vertical-domain, and faithfulness benchmarks. Our training code and models are coming soon.

cs.CL↗

arXiv2026

Observation of ΔJ=0 Rotational Excitation in Dense Hydrogens

Raman measurements performed on dense H2, D2 and H2+D2 in a wide pressure-temperature range reveal the presence of the ΔJ=0 rotational excitation. In the gas/fluid state this excitation has zero Raman shift, but in the solid, the crystal field drive s it away from the zero value e.g. 75 cm-1 at around 50 GPa and 10 K for both isotopes and their mixture. In the case of deuterium, the ΔJ=0 mode splits upon entering phase II suggesting a very complex molecular environment of the broken symmetry phase (BSP). In the fluid state and phases I and II the frequencies (energies) of the ΔJ=0 transition for H2 and D2 do not scale either as rotational (by factor of 2) nor vibrational (by square 2) modes and appear to be completely isotope independent. This independence on mass marks this transition as unique and a fundamentally different type of excitation from the commonly considered harmonic oscillator and quantum rotor.

quant-ph↗

arXiv2026

Tabletop X-ray ghost video of moving objects

X-ray imaging is widely employed in clinical medicine, industrial inspection, and various scientific research fields. Unfortunately, most currently used X-ray two-dimensional (2D) detectors suffer from a fundamental trade-off between the number of pixels and readout time, making them unsuitable for fast moving objects imaging, as well as the readout dead time causes frame losses. X-ray ghost imaging (XGI) offers an alternative approach to image an object using only a highly sensitive single-pixel detector. However, a critical limitation of existing XGI methods is the excessive total acquisition time required, rendering it impractical for real applications. In this paper, we propose a rapid spatial modulation scheme based on random binary patterns encoded onto a fast-spinning mask. Clear X-ray visualization of moving objects is demonstrated with imaging rates up to 200 frames per second with a resolution of 225 um. For the first time, our method has greatly improved the XGI imaging speed and paves the way for X-ray imaging application of motion objects, such as the inspection of rotating aero-engines and in vivo medical imaging.

physics.optics↗

arXiv2025

Breaking Data Silos: Towards Open and Scalable Mobility Foundation Models via Generative Continual Learning

Human mobility is a fundamental pillar of urban science and sustainability, providing critical insights into energy consumption, carbon emissions, and public health. However, the discovery of universal mobility laws is currently hindered by the ``data silo'' problem, where institutional boundaries and privacy regulations fragment the necessary large-scale datasets. In this paper, we propose MoveGCL, a transformative framework that facilitates collaborative and decentralized mobility science via generative continual learning. MoveGCL enables a distributed ecosystem of data holders to jointly evolve a foundation model without compromising individual privacy. The core of MoveGCL lies in its ability to replay synthetic trajectories derived from a generative teacher and utilize a mobility-pattern-aware Mixture-of-Experts (MoE) architecture. This allows the model to encapsulate the unique characteristics of diverse urban structures while mitigating the risk of knowledge erosion (catastrophic forgetting). With a specialized layer-wise progressive adaptation strategy, MoveGCL ensures stable convergence during the continuous integration of new urban domains. Our experiments on six global urban datasets demonstrate that MoveGCL achieves performance parity with joint training, a previously unattainable feat under siloed conditions. This work provides a scalable, privacy-preserving pathway toward Open Mobility Science, empowering researchers to address global sustainability challenges through cross-institutional AI collaboration. To facilitate reproducibility and future research, we have released the code and models at \color{blue}{https://github.com/tsinghua-fib-lab/MoveGCL}.

cs.LG↗

arXiv2025

TrafficSimAgent: A Hierarchical Agent Framework for Autonomous Traffic Simulation with MCP Control

Traffic simulation is important for transportation optimization and policy making. While existing simulators such as SUMO and MATSim offer fully-featured platforms and utilities, users without too much knowledge about these platforms often face significant challenges when conducting experiments from scratch and applying them to their daily work. To solve this challenge, we propose TrafficSimAgent, an LLM-based agent framework that serves as an expert in experiment design and decision optimization for general-purpose traffic simulation tasks. The framework facilitates execution through cross-level collaboration among expert agents: high-level expert agents comprehend natural language instructions with high flexibility, plan the overall experiment workflow, and invoke corresponding MCP-compatible tools on demand; meanwhile, low-level expert agents select optimal action plans for fundamental elements based on real-time traffic conditions. Extensive experiments across multiple scenarios show that TrafficSimAgent effectively executes simulations under various conditions and consistently produces reasonable outcomes even when user instructions are ambiguous. Besides, the carefully designed expert-level autonomous decision-driven optimization in TrafficSimAgent yields superior performance when compared with other systems and SOTA LLM based methods.

cs.AI↗

arXiv2025

FutureX: Enhance End-to-End Autonomous Driving via Latent Chain-of-Thought World Model

In autonomous driving, end-to-end planners learn scene representations from raw sensor data and utilize them to generate a motion plan or control actions. However, exclusive reliance on the current scene for motion planning may result in suboptimal responses in highly dynamic traffic environments where ego actions further alter the future scene. To model the evolution of future scenes, we leverage the World Model to represent how the ego vehicle and its environment interact and change over time, which entails complex reasoning. The Chain of Thought (CoT) offers a promising solution by forecasting a sequence of future thoughts that subsequently guide trajectory refinement. In this paper, we propose FutureX, a CoT-driven pipeline that enhances end-to-end planners to perform complex motion planning via future scene latent reasoning and trajectory refinement. Specifically, the Auto-think Switch examines the current scene and decides whether additional reasoning is required to yield a higher-quality motion plan. Once FutureX enters the Thinking mode, the Latent World Model conducts a CoT-guided rollout to predict future scene representation, enabling the Summarizer Module to further refine the motion plan. Otherwise, FutureX operates in an Instant mode to generate motion plans in a forward pass for relatively simple scenes. Extensive experiments demonstrate that FutureX enhances existing methods by producing more rational motion plans and fewer collisions without compromising efficiency, thereby achieving substantial overall performance gains, e.g., 6.2 PDMS improvement for TransFuser on NAVSIM. Code will be released.

cs.CV↗

arXiv2025

Soft-Lubrication Drainage and Rupture in Particle-Driven Vesicles

The deformation and rupture of a lipid vesicle due to the forced normal approach of an inclusion are essential for optimizing the design of magnetic giant unilamellar vesicles [magGUVs, Malik et al., Nanoscale 17, 13720 (2025)], with implications for active colloid-membrane interactions and cellular-scale chemical delivery. Here, we investigate vesicles propelled by a force-driven rigid inclusion and reveal a robust elastohydrodynamic mechanism: the inclusion outpaces the vesicle, sustaining a thinning film that drains symmetrically and self-similarly, largely independent of initial shape. For soft membranes and small inclusions, coupling drives a monotonic tension increase that can exceed the lysis tension. Evaluating the maximal tension over a delivery distance, we map an operating window in vesicle reduced area and size relative to the inclusion.

cond-mat.soft↗

arXiv2025

Understanding World or Predicting Future? A Comprehensive Survey of World Models

The concept of world models has garnered significant attention due to advancements in multimodal large language models such as GPT-4 and video generation models such as Sora, which are central to the pursuit of artificial general intelligence. This survey offers a comprehensive review of the literature on world models. Generally, world models are regarded as tools for either understanding the present state of the world or predicting its future dynamics. This review presents a systematic categorization of world models, emphasizing two primary functions: (1) constructing internal representations to understand the mechanisms of the world, and (2) predicting future states to simulate and guide decision-making. Initially, we examine the current progress in these two categories. We then explore the application of world models in key domains, including generative games, autonomous driving, robotics, and social simulacra, with a focus on how each domain utilizes these aspects. Finally, we outline key challenges and provide insights into potential future research directions. We summarize the representative papers along with their code repositories in https://github.com/tsinghua-fib-lab/World-Model.

cs.CL↗

arXiv2025

AR-Med: Automated Relevance Enhancement in Medical Search via LLM-Driven Information Augmentation

Accurate and reliable search on online healthcare platforms is critical for user safety and service efficacy. Traditional methods, however, often fail to comprehend complex and nuanced user queries, limiting their effectiveness. Large language models (LLMs) present a promising solution, offering powerful semantic understanding to bridge this gap. Despite their potential, deploying LLMs in this high-stakes domain is fraught with challenges, including factual hallucinations, specialized knowledge gaps, and high operational costs. To overcome these barriers, we introduce \textbf{AR-Med}, a novel framework for \textbf{A}utomated \textbf{R}elevance assessment for \textbf{Med}ical search that has been successfully deployed at scale on the Online Medical Delivery Platforms. AR-Med grounds LLM reasoning in verified medical knowledge through a retrieval-augmented approach, ensuring high accuracy and reliability. To enable efficient online service, we design a practical knowledge distillation scheme that compresses large teacher models into compact yet powerful student models. We also introduce LocalQSMed, a multi-expert annotated benchmark developed to guide model iteration and ensure strong alignment between offline and online performance. Extensive experiments show AR-Med achieves an offline accuracy of over 93\%, a 24\% absolute improvement over the original online system, and delivers significant gains in online relevance and user satisfaction. Our work presents a practical and scalable blueprint for developing trustworthy, LLM-powered systems in real-world healthcare applications.

cs.CL↗

arXiv2025

MoveGPT: Scaling Mobility Foundation Models with Spatially-Aware Mixture of Experts

The success of foundation models in language has inspired a new wave of general-purpose models for human mobility. However, existing approaches struggle to scale effectively due to two fundamental limitations: a failure to use meaningful basic units to represent movement, and an inability to capture the vast diversity of patterns found in large-scale data. In this work, we develop MoveGPT, a large-scale foundation model specifically architected to overcome these barriers. MoveGPT is built upon two key innovations: (1) a unified location encoder that maps geographically disjoint locations into a shared semantic space, enabling pre-training on a global scale; and (2) a Spatially-Aware Mixture-of-Experts Transformer that develops specialized experts to efficiently capture diverse mobility patterns. Pre-trained on billion-scale datasets, MoveGPT establishes a new state-of-the-art across a wide range of downstream tasks, achieving performance gains of up to 35% on average. It also demonstrates strong generalization capabilities to unseen cities. Crucially, our work provides empirical evidence of scaling ability in human mobility, validating a clear path toward building increasingly capable foundation models in this domain.

cs.AI↗