arXiv subjects

Jie Yang

Publications and source records attributed to Jie Yang.

At least 199 records · Page 11Linked to original sources

arXiv2025

Lunar and Terrestrial Time Transformation Based on the Principle of General Relativity

Lunar time metrology necessitates a unified temporal framework beyond Earth, requiring an independent lunar system for timekeeping, dissemination, and calendrics. Recent American publications define Lunar Coordinate Time (LTC) within relativity and propose a Terrestrial Time (TT) to LTC conversion formula. However, this formula's derivation and assumptions are contested. The complex dynamics within the solar system can be simplified by decomposing relationships into hierarchical wide-area (external problem) and local-area (internal problem) levels. Grounded in the symmetry and conservation laws of physics, Einstein's general relativity emphasizes two key principles: (i) Equal weighting: Relationships among multi-level coordinate systems are independent and self-similar (analogous to fractals). (ii) *Locality*: The laws of physics retain invariant forms only in local coordinate systems. Specifically, a non-rotating system corresponds to the Frenet frame along a particle's geodesic. Preserving physical law invariance requires restricting rotating references strictly to the local domain; defining the orientation of an Earth-centered system using distant celestial bodies violates general relativity's locality principle. This work derives the relationship between coordinate time and proper time. Using the Earth-Moon system as an intermediary, it obtains a simplified transformation formula between LTC and TT. An independent and universal lunar standard time framework is proposed. Crucially, the derived coordinate time transformation coefficient exhibits long-term secular variation. This variation can be measured and predicted through precise Earth-Moon time comparisons.

gr-qc↗

arXiv2025

CD-NGP: A Fast Scalable Continual Representation for Dynamic Scenes

Novel view synthesis (NVS) in dynamic scenes faces persistent challenges in memory consumption, model complexity, training efficiency, and rendering quality. Offline methods offer high fidelity but suffer from high memory usage and limited scalability, while online approaches often trade quality for speed and compactness. We propose Continual Dynamic Neural Graphics Primitives (CD-NGP), a continual learning framework that reduces memory overhead and enhances scalability through parameter reuse. To avoid feature interference in dynamic scenes and improve rendering quality, our method combines spatial and temporal hash encodings, which compactly represent scene structures and motion patterns. We also introduce a new dataset comprising multi-view, long-duration ($>1200$ frames) videos with both rigid and non-rigid motion, which is not found in existing benchmarks. CD-NGP is evaluated on public datasets and our long video dataset, demonstrating superior scalability and reconstruction quality. It significantly reduces training memory usage to <14GB and requires only 0.4MB/frame in streaming bandwidth on DyNeRF -- substantially lower than most online baselines.

cs.CV↗

arXiv2025

De novo design of alpha-helical peptide amphiphiles repairing fragmented collagen type I via supramolecular co-assembly

The hierarchical triple-helix structure of collagen type I, Col I, is essential for extracellular matrix support and integrity. However, current reconstruction strategies face challenges such as chain mismatch, preventing proper fibril formation. Here, we report a supramolecular co-assembly strategy using a de novo-designed alpha-helical peptide amphiphile (APA) of just seven amino acids. The APA features a hydrophobic palmitic acid tail, which stabilizes the helical structure and promotes co-assembly upon interaction with complementary molecular structures. This minimal design enables selective recognition of fragmented collagen (FC), restoring triple-helix conformation and guiding fibre formation. We applied this mechanism to engineer FC-rich nanofat (NF) into a mechanically reinforced biomaterial. Integration of APA-NF with coaxial 3D printing enabled spatial control of structure and function. In a porcine model, this platform enhanced in situ vascularized adipose tissue regeneration. Our results demonstrate that hierarchical reconstruction of collagen via peptide-guided supramolecular assembly offers a promising strategy for soft tissue repair.

physics.chem-ph↗

arXiv2025

Style over Substance: Distilled Language Models Reason Via Stylistic Replication

Specialized reasoning language models (RLMs) have demonstrated that scaling test-time computation through detailed reasoning traces significantly enhances performance. Although these traces effectively facilitate knowledge distillation into smaller, instruction-tuned models, the precise nature of transferred reasoning remains unclear. In this study, we investigate to what extent distilled models internalize replicated stylistic patterns during reasoning. To this end, we systematically analyze reasoning traces, identifying structural and lexical patterns that characterize successful reasoning. We then introduce two new datasets -- a dataset of emergent reasoning traces and a synthetic dataset explicitly constructed to replicate these stylistic patterns -- to precisely examine their influence on distilled models' reasoning capabilities. We find that models trained on the synthetic traces achieve comparable performance, indicating that distilled reasoning abilities rely significantly on surface-level patterns. Surprisingly, we observe an increase in performance even when the synthetic traces are altered to lead to the wrong answer. Our findings highlight how stylistic patterns can be leveraged to efficiently enhance LM reasoning across diverse model families.

cs.CL↗

arXiv2025

KptLLM++: Towards Generic Keypoint Comprehension with Large Language Model

The emergence of Multimodal Large Language Models (MLLMs) has revolutionized image understanding by bridging textual and visual modalities. However, these models often struggle with capturing fine-grained semantic information, such as the precise identification and analysis of object keypoints. Keypoints, as structure-aware, pixel-level, and compact representations of objects, particularly articulated ones, play a crucial role in applications such as fine-grained image analysis, object retrieval, and behavior recognition. In this paper, we propose KptLLM++, a novel multimodal large language model that specifically designed for generic keypoint comprehension through the integration of diverse input modalities guided by user-defined instructions. By unifying keypoint detection across varied contexts, KptLLM++ establishes itself as an advanced interface, fostering more effective human-AI collaboration. The model is built upon a novel identify-then-detect paradigm, which first interprets keypoint semantics and subsequently localizes their precise positions through a structured chain-of-thought reasoning mechanism. To push the boundaries of performance, we have scaled up the training dataset to over 500K samples, encompassing diverse objects, keypoint categories, image styles, and scenarios with complex occlusions. This extensive scaling enables KptLLM++ to unlock its potential, achieving remarkable accuracy and generalization. Comprehensive experiments on multiple keypoint detection benchmarks demonstrate its state-of-the-art performance, underscoring its potential as a unified solution for fine-grained image understanding and its transformative implications for human-AI interaction.

cs.CV↗

arXiv2025

Multinomial Link Models

We propose a new family of regression models for analyzing categorical responses, called multinomial link models. It consists of four classes, namely, mixed-link models that generalize existing multinomial logistic models and their extensions, two-group models that can incorporate the observations with NA or unknown responses, dichotomous conditional link models that handle longitudinal binary responses, and po-npo mixture models that are more flexible than partial proportional odds models. By characterizing the feasible parameter space, deriving necessary and sufficient conditions, and developing validated algorithms to guarantee the finding of feasible maximum likelihood estimates, we solve the infeasibility issue of existing statistical software when estimating parameters for cumulative link models. We also provide explicit formulae and detailed algorithms for computing the Fisher information matrix and selecting the best models among the new family. The applications to real datasets show that the new models can fit the data significantly better, correct misleading conclusions due to missing responses, and make more informative statistical inference.

stat.ME↗

arXiv2025

Cooperative Mapping, Localization, and Beam Management via Multi-Modal SLAM in ISAC Systems

Simultaneous localization and mapping (SLAM) plays a critical role in integrated sensing and communication (ISAC) systems for sixth-generation (6G) millimeter-wave (mmWave) networks, enabling environmental awareness and precise user equipment (UE) positioning. While cooperative multi-user SLAM has demonstrated potential in leveraging distributed sensing, its application within multi-modal ISAC systems remains limited, particularly in terms of theoretical modeling and communication-layer integration. This paper proposes a novel multi-modal SLAM framework that addresses these limitations through three key contributions. First, a Bayesian estimation framework is developed for cooperative multi-user SLAM, along with a two-stage algorithm for robust radio map construction under dynamic and heterogeneous sensing conditions. Second, a multi-modal localization strategy is introduced, fusing SLAM results with camera-based multi-object tracking and inertial measurement unit (IMU) data via an error-aware model, significantly improving UE localization in multi-user scenarios. Third, a sensing-aided beam management scheme is proposed, utilizing global radio maps and localization data to generate UE-specific prior information for beam selection, thereby reducing inter-user interference and enhancing downlink spectral efficiency. Simulation results demonstrate that the proposed system improves radio map accuracy by up to 60%, enhances localization accuracy by 37.5%, and significantly outperforms traditional methods in both indoor and outdoor environments.

cs.IT↗

arXiv2025

Simultaneous Localization and Mapping Using Active mmWave Sensing in 5G NR

Millimeter-wave (mmWave) 5G New Radio (NR) communication systems, with their high-resolution antenna arrays and extensive bandwidth, offer a transformative opportunity for high-throughput data transmission and advanced environmental sensing. Although passive sensing-based SLAM techniques can estimate user locations and environmental reflections simultaneously, their effectiveness is often constrained by assumptions of specular reflections and oversimplified map representations. To overcome these limitations, this work employs a mmWave 5G NR system for active sensing, enabling it to function similarly to a laser scanner for point cloud generation. Specifically, point clouds are extracted from the power delay profile estimated from each beam direction using a binary search approach. To ensure accuracy, hardware delays are calibrated with multiple predefined target points. Pose variations of the terminal are then estimated from point cloud data gathered along continuous trajectory viewpoints using point cloud registration algorithms. Loop closure detection and pose graph optimization are subsequently applied to refine the sensing results, achieving precise terminal localization and detailed radio map reconstruction. The system is implemented and validated through both simulations and experiments, confirming the effectiveness of the proposed approach.

eess.SP↗

arXiv2025

FB-Diff: Fourier Basis-guided Diffusion for Temporal Interpolation of 4D Medical Imaging

The temporal interpolation task for 4D medical imaging, plays a crucial role in clinical practice of respiratory motion modeling. Following the simplified linear-motion hypothesis, existing approaches adopt optical flow-based models to interpolate intermediate frames. However, realistic respiratory motions should be nonlinear and quasi-periodic with specific frequencies. Intuited by this property, we resolve the temporal interpolation task from the frequency perspective, and propose a Fourier basis-guided Diffusion model, termed FB-Diff. Specifically, due to the regular motion discipline of respiration, physiological motion priors are introduced to describe general characteristics of temporal data distributions. Then a Fourier motion operator is elaborately devised to extract Fourier bases by incorporating physiological motion priors and case-specific spectral information in the feature space of Variational Autoencoder. Well-learned Fourier bases can better simulate respiratory motions with motion patterns of specific frequencies. Conditioned on starting and ending frames, the diffusion model further leverages well-learned Fourier bases via the basis interaction operator, which promotes the temporal interpolation task in a generative manner. Extensive results demonstrate that FB-Diff achieves state-of-the-art (SOTA) perceptual performance with better temporal consistency while maintaining promising reconstruction metrics. Codes are available.

eess.IV↗

arXiv2025

Difference Imaging-Based Parking Lot Surveillance in Multi-RIS-Aided Collaborative ISAC System

Parking lot surveillance with integrated sensing and communication (ISAC) system is one of the potential application scenarios defined by 3rd Generation Partnership Project (3GPP). Traditional surveillance systems using cameras or magnetic sensors face limitations such as light dependence, high costs, and constrained scalability. Wireless sensing with reconfigurable intelligent surfaces (RISs) has the ability to address the above limitations due to its light independence and lower deployment overhead. In this study, we propose a difference imaging-based multi-RIS-aided collaborative ISAC system to achieve parking lot surveillance. In a parking lot, the presence of vehicles induces impacts on wireless environments due to scattering characteristic variation. By delineating the parking lot into a two-dimensional image with several grid units, the proposed system can capture the variation of their scattering coefficients in free and occupied states. The variation between these two states is sparse, which can be captured through compressed sensing (CS)-based imaging algorithms. Additionally, we collaboratively employ multiple RISs to enable higher surveillance performance. Experimental results demonstrate that our method can achieve high-accuracy parking occupancy detection, and the employment of collaborative RISs further enhances the detection rate.

cs.IT↗

arXiv2025

RIS-Aided Cooperative ISAC Networks for Structural Health Monitoring

Integrated sensing and communication (ISAC) is a key feature of future cellular systems, enabling applications such as intruder detection, monitoring, and tracking using the same infrastructure. However, its potential for structural health monitoring (SHM), which requires the detection of slow and subtle structural changes, remains largely unexplored due to challenges such as multipath interference and the need for ultra-high sensing precision. This study introduces a novel theoretical framework for SHM via ISAC by leveraging reconfigurable intelligent surfaces (RIS) as reference points in collaboration with base stations and users. By dynamically adjusting RIS phases to generate distinct radio signals that suppress background multipath interference, measurement accuracy at these reference points is enhanced. We theoretically analyze RIS-aided collaborative sensing in three-dimensional cellular networks using Fisher information theory, demonstrating how increasing observation time, incorporating additional receivers (even with self-positioning errors), optimizing RIS phases, and refining collaborative node selection can reduce the position error bound to meet SHM's stringent accuracy requirements. Furthermore, we develop a Bayesian inference model to identify structural states and validate damage detection probabilities. Both theoretical and numerical analyses confirm ISAC's capability for millimeter-level deformation detection, highlighting its potential for high-precision SHM applications.

cs.IT↗

arXiv2025

Zero-Shot Contextual Embeddings via Offline Synthetic Corpus Generation

Context-aware embedding methods boost retrieval accuracy by conditioning on corpus statistics (e.g., term co-occurrence and topical patterns) extracted from neighboring documents. However, this context-aware approach requires access to the target corpus or requires domain-specific finetuning, posing practical barriers in privacy-sensitive or resource-constrained settings. We present ZEST, a zero-shot contextual adaptation framework that replaces real corpus access with a one-time offline synthesis of a compact proxy. Given only a handful exemplar documents representative of the general target domain, we use a multi-step hierarchical procedure to generate a synthetic context corpus of several hundred documents that aims to emulate key domain-specific distributions. At inference, the frozen context-aware encoder uses this proxy corpus -- without any finetuning or target corpus access -- to produce domain-adapted embeddings. Across the MTEB benchmark, ZEST's zero-shot synthetic context adaptation using only five example documents performs within 0.5% of models leveraging full target corpus access -- demonstrating remarkable efficacy without any retraining. ZEST thus provides a practical method for deploying high-performance, adaptable embeddings in constrained environments.

cs.CL↗

arXiv2025

Temporal Differential Fields for 4D Motion Modeling via Image-to-Video Synthesis

Temporal modeling on regular respiration-induced motions is crucial to image-guided clinical applications. Existing methods cannot simulate temporal motions unless high-dose imaging scans including starting and ending frames exist simultaneously. However, in the preoperative data acquisition stage, the slight movement of patients may result in dynamic backgrounds between the first and last frames in a respiratory period. This additional deviation can hardly be removed by image registration, thus affecting the temporal modeling. To address that limitation, we pioneeringly simulate the regular motion process via the image-to-video (I2V) synthesis framework, which animates with the first frame to forecast future frames of a given length. Besides, to promote the temporal consistency of animated videos, we devise the Temporal Differential Diffusion Model to generate temporal differential fields, which measure the relative differential representations between adjacent frames. The prompt attention layer is devised for fine-grained differential fields, and the field augmented layer is adopted to better interact these fields with the I2V framework, promoting more accurate temporal variation of synthesized videos. Extensive results on ACDC cardiac and 4D Lung datasets reveal that our approach simulates 4D videos along the intrinsic motion trajectory, rivaling other competitive methods on perceptual similarity and temporal consistency. Codes will be available soon.

cs.CV↗

arXiv2025

CAD-GPT: Synthesising CAD Construction Sequence with Spatial Reasoning-Enhanced Multimodal LLMs

Computer-aided design (CAD) significantly enhances the efficiency, accuracy, and innovation of design processes by enabling precise 2D and 3D modeling, extensive analysis, and optimization. Existing methods for creating CAD models rely on latent vectors or point clouds, which are difficult to obtain, and storage costs are substantial. Recent advances in Multimodal Large Language Models (MLLMs) have inspired researchers to use natural language instructions and images for CAD model construction. However, these models still struggle with inferring accurate 3D spatial location and orientation, leading to inaccuracies in determining the spatial 3D starting points and extrusion directions for constructing geometries. This work introduces CAD-GPT, a CAD synthesis method with spatial reasoning-enhanced MLLM that takes either a single image or a textual description as input. To achieve precise spatial inference, our approach introduces a 3D Modeling Spatial Mechanism. This method maps 3D spatial positions and 3D sketch plane rotation angles into a 1D linguistic feature space using a specialized spatial unfolding mechanism, while discretizing 2D sketch coordinates into an appropriate planar space to enable precise determination of spatial starting position, sketch orientation, and 2D sketch coordinate translations. Extensive experiments demonstrate that CAD-GPT consistently outperforms existing state-of-the-art methods in CAD model synthesis, both quantitatively and qualitatively.

cs.CV↗

arXiv2025

Step-Audio-AQAA: a Fully End-to-End Expressive Large Audio Language Model

Large Audio-Language Models (LALMs) have significantly advanced intelligent human-computer interaction, yet their reliance on text-based outputs limits their ability to generate natural speech responses directly, hindering seamless audio interactions. To address this, we introduce Step-Audio-AQAA, a fully end-to-end LALM designed for Audio Query-Audio Answer (AQAA) tasks. The model integrates a dual-codebook audio tokenizer for linguistic and semantic feature extraction, a 130-billion-parameter backbone LLM and a neural vocoder for high-fidelity speech synthesis. Our post-training approach employs interleaved token-output of text and audio to enhance semantic coherence and combines Direct Preference Optimization (DPO) with model merge to improve performance. Evaluations on the StepEval-Audio-360 benchmark demonstrate that Step-Audio-AQAA excels especially in speech control, outperforming the state-of-art LALMs in key areas. This work contributes a promising solution for end-to-end LALMs and highlights the critical role of token-based vocoder in enhancing overall performance for AQAA tasks.

cs.SD↗

arXiv2025

Learned Off-Grid Imager for Low-Altitude Economy with Cooperative ISAC Network

The low-altitude economy is emerging as a key driver of future economic growth, necessitating effective flight activity surveillance using existing mobile cellular network sensing capabilities. However, traditional monostatic and localizationbased sensing methods face challenges in fusing sensing results and matching channel parameters. To address these challenges, we model low-altitude surveillance as a compressed sensing (CS)-based imaging problem by leveraging the cooperation of multiple base stations and the inherent sparsity of aerial images. Additionally, we derive the point spread function to analyze the influences of different antenna, subcarrier, and resolution settings on the imaging performance. Given the random spatial distribution of unmanned aerial vehicles (UAVs), we propose a physics-embedded learning method to mitigate off-grid errors in traditional CS-based approaches. Furthermore, to enhance rare UAV detection in vast low-altitude airspace, we integrate an online hard example mining scheme into the loss function design, enabling the network to adaptively focus on samples with significant discrepancies from the ground truth during training. Simulation results demonstrate the effectiveness of the proposed low-altitude surveillance framework. The proposed physicsembedded learning algorithm achieves a 97.55% detection rate, significantly outperforming traditional CS-based methods under off-grid conditions. Part of the source code for this paper will be soon accessed at https://github.com/kiwi1944/LAEImager.

cs.IT↗

arXiv2025

WeThink: Toward General-purpose Vision-Language Reasoning via Reinforcement Learning

Building on the success of text-based reasoning models like DeepSeek-R1, extending these capabilities to multimodal reasoning holds great promise. While recent works have attempted to adapt DeepSeek-R1-style reinforcement learning (RL) training paradigms to multimodal large language models (MLLM), focusing on domain-specific tasks like math and visual perception, a critical question remains: How can we achieve the general-purpose visual-language reasoning through RL? To address this challenge, we make three key efforts: (1) A novel Scalable Multimodal QA Synthesis pipeline that autonomously generates context-aware, reasoning-centric question-answer (QA) pairs directly from the given images. (2) The open-source WeThink dataset containing over 120K multimodal QA pairs with annotated reasoning paths, curated from 18 diverse dataset sources and covering various question domains. (3) A comprehensive exploration of RL on our dataset, incorporating a hybrid reward mechanism that combines rule-based verification with model-based assessment to optimize RL training efficiency across various task domains. Across 14 diverse MLLM benchmarks, we demonstrate that our WeThink dataset significantly enhances performance, from mathematical reasoning to diverse general multimodal tasks. Moreover, we show that our automated data pipeline can continuously increase data diversity to further improve model performance.

cs.CV↗

arXiv2025

InteractAnything: Zero-shot Human Object Interaction Synthesis via LLM Feedback and Object Affordance Parsing

Recent advances in 3D human-aware generation have made significant progress. However, existing methods still struggle with generating novel Human Object Interaction (HOI) from text, particularly for open-set objects. We identify three main challenges of this task: precise human-object relation reasoning, affordance parsing for any object, and detailed human interaction pose synthesis aligning description and object geometry. In this work, we propose a novel zero-shot 3D HOI generation framework without training on specific datasets, leveraging the knowledge from large-scale pre-trained models. Specifically, the human-object relations are inferred from large language models (LLMs) to initialize object properties and guide the optimization process. Then we utilize a pre-trained 2D image diffusion model to parse unseen objects and extract contact points, avoiding the limitations imposed by existing 3D asset knowledge. The initial human pose is generated by sampling multiple hypotheses through multi-view SDS based on the input text and object geometry. Finally, we introduce a detailed optimization to generate fine-grained, precise, and natural interaction, enforcing realistic 3D contact between the 3D object and the involved body parts, including hands in grasping. This is achieved by distilling human-level feedback from LLMs to capture detailed human-object relations from the text instruction. Extensive experiments validate the effectiveness of our approach compared to prior works, particularly in terms of the fine-grained nature of interactions and the ability to handle open-set 3D objects.

cs.CV↗