arXiv subjects

Jie Jiang

Publications and source records attributed to Jie Jiang.

At least 145 records · Page 8Linked to original sources

arXiv2025

HunyuanVideo: A Systematic Framework For Large Video Generative Models

Recent advancements in video generation have significantly impacted daily life for both individuals and industries. However, the leading video generation models remain closed-source, resulting in a notable performance gap between industry capabilities and those available to the public. In this report, we introduce HunyuanVideo, an innovative open-source video foundation model that demonstrates performance in video generation comparable to, or even surpassing, that of leading closed-source models. HunyuanVideo encompasses a comprehensive framework that integrates several key elements, including data curation, advanced architectural design, progressive model scaling and training, and an efficient infrastructure tailored for large-scale model training and inference. As a result, we successfully trained a video generative model with over 13 billion parameters, making it the largest among all open-source models. We conducted extensive experiments and implemented a series of targeted designs to ensure high visual quality, motion dynamics, text-video alignment, and advanced filming techniques. According to evaluations by professionals, HunyuanVideo outperforms previous state-of-the-art models, including Runway Gen-3, Luma 1.6, and three top-performing Chinese video generative models. By releasing the code for the foundation model and its applications, we aim to bridge the gap between closed-source and open-source communities. This initiative will empower individuals within the community to experiment with their ideas, fostering a more dynamic and vibrant video generation ecosystem. The code is publicly available at https://github.com/Tencent/HunyuanVideo.

cs.CV↗

arXiv2025

Flexible Exoskeleton Control Based on Binding Alignment Strategy and Full-arm Coordination Mechanism

In rehabilitation, powered, and teleoperation exoskeletons, connecting the human body to the exoskeleton through binding attachments is a common configuration. However, the uncertainty of the tightness and the donning deviation of the binding attachments will affect the flexibility and comfort of the exoskeletons, especially during high-speed movement. To address this challenge, this paper presents a flexible exoskeleton control approach with binding alignment and full-arm coordination. Firstly, the sources of the force interaction caused by donning offsets are analyzed, based on which the interactive force data is classified into the major, assistant, coordination, and redundant component categories. Then, a binding alignment strategy (BAS) is proposed to reduce the donning disturbances by combining different force data. Furthermore, we propose a full-arm coordination mechanism (FCM) that focuses on two modes of arm movement intent, joint-oriented and target-oriented, to improve the flexible performance of the whole exoskeleton control during high-speed motion. In this method, we propose an algorithm to distinguish the two intentions to resolve the conflict issue of the force component. Finally, a series of experiments covering various aspects of exoskeleton performance (flexibility, adaptability, accuracy, speed, and fatigue) were conducted to demonstrate the benefits of our control framework in our full-arm exoskeleton.

cs.RO↗

arXiv2025

Mutual Validation of Datasets for Analyzing Tilt Angles in Solar Active Regions

The tilt angle of solar active regions (AR) is crucial for the Babcock-Leighton type dynamo models in the buildup of polar field. However, divergent results regarding properties of tilt angles were reported due to their wide scatter, caused by intrinsic solar mechanisms and measurement errors. Here, we mutually validate the magnetogram-based AR tilt angle dataset from Wang, Jiang, & Luo with the Debrecen Photoheliographic Data by identifying common data points where both datasets provide comparable tilt angles for the same AR/sunspot. The mutually validated datasets effectively reduce measurement errors, enabling a more accurate analysis of the intrinsic properties of tilt angles. Our mutually validated datasets reveal that the difference between white-light-based and magnetogram-based tilt angles has no significant difference. Also, the datasets show that an upward revision of average tilt angle ($\barα$) and a downward revision of the tilt scatter ($σ_α$) compared to previous results are necessary, with typical values of about 7$^\circ$ and 16$^\circ$, respectively. The $σ_α$ values demonstrate a strong correlation with AR flux and sunspot area, with the dependency functions re-evaluated using mutually validated datasets. Furthermore, both $\barα$ and the tilt coefficient for the weak cycle 24 are larger than those for cycle 23. This supports the tilt quenching mechanism, which posits an anti-correlation between the cycle-averaged tilt angle and cycle amplitude. Additionally, tilt angle from the mutually validated dataset has a weak non-monotonic relationship with magnetic flux and does not depend on the maximum magnetic field strength of ARs.

astro-ph.SR↗

arXiv2025

Anomalous Raman scattering in layered AgCrP$_2$Se$_6$: Helical modes and excitation energy-dependent intensities

Structural anisotropy in layered two-dimensional materials can lead to highly anisotropic optical absorption which, in turn, can profoundly a^ect their phonon modes. These e^ects include lattice orientation-dependent and excitation energy-dependent mode intensities that can enable new phononic and optoelectronic applications. Here, we report anomalous Raman spectra in single-crystalline AgCrP$_2$Se$_6$, a layered antiferromagnetic material. Density functional theory calculations and experimental measurements reveal several unique features in the Raman spectra of bulk and exfoliated AgCrP$_2$Se$_6$ crystals including three helical vibrational modes. These modes exhibit large Raman optical activities (circular intensity di^erences) in bulk AgCrP$_2$Se$_6$, which progressively decrease with thickness. We also observe strong excitation energy dependent peak intensities as well as a decrease in anti-Stokes peak intensities at room temperature with increasing excitation energy, resulting in an apparent cooling by up to 220 K. All of these anomalies in bulk and exfoliated flakes are attributed to the unique ABC layer stacking structure of AgCrP$_2$Se$_6$ and to the smaller unit cell volume that causes hybridization between the Se and Ag/Cr electron densities, resulting in charge transfer and strongly a^ecting the electron-phonon coupling. This work thus positions AgCrP$_2$Se$_6$ as an exciting new 2D material for optical and phononic applications.

cond-mat.mes-hall↗

arXiv2025

Hunyuan3D 1.0: A Unified Framework for Text-to-3D and Image-to-3D Generation

While 3D generative models have greatly improved artists' workflows, the existing diffusion models for 3D generation suffer from slow generation and poor generalization. To address this issue, we propose a two-stage approach named Hunyuan3D 1.0 including a lite version and a standard version, that both support text- and image-conditioned generation. In the first stage, we employ a multi-view diffusion model that efficiently generates multi-view RGB in approximately 4 seconds. These multi-view images capture rich details of the 3D asset from different viewpoints, relaxing the tasks from single-view to multi-view reconstruction. In the second stage, we introduce a feed-forward reconstruction model that rapidly and faithfully reconstructs the 3D asset given the generated multi-view images in approximately 7 seconds. The reconstruction network learns to handle noises and in-consistency introduced by the multi-view diffusion and leverages the available information from the condition image to efficiently recover the 3D structure. Our framework involves the text-to-image model, i.e., Hunyuan-DiT, making it a unified framework to support both text- and image-conditioned 3D generation. Our standard version has 3x more parameters than our lite and other existing model. Our Hunyuan3D 1.0 achieves an impressive balance between speed and quality, significantly reducing generation time while maintaining the quality and diversity of the produced assets.

cs.CV↗

arXiv2025

MedFILIP: Medical Fine-grained Language-Image Pre-training

Medical vision-language pretraining (VLP) that leverages naturally-paired medical image-report data is crucial for medical image analysis. However, existing methods struggle to accurately characterize associations between images and diseases, leading to inaccurate or incomplete diagnostic results. In this work, we propose MedFILIP, a fine-grained VLP model, introduces medical image-specific knowledge through contrastive learning, specifically: 1) An information extractor based on a large language model is proposed to decouple comprehensive disease details from reports, which excels in extracting disease deals through flexible prompt engineering, thereby effectively reducing text complexity while retaining rich information at a tiny cost. 2) A knowledge injector is proposed to construct relationships between categories and visual attributes, which help the model to make judgments based on image features, and fosters knowledge extrapolation to unfamiliar disease categories. 3) A semantic similarity matrix based on fine-grained annotations is proposed, providing smoother, information-richer labels, thus allowing fine-grained image-text alignment. 4) We validate MedFILIP on numerous datasets, e.g., RSNA-Pneumonia, NIH ChestX-ray14, VinBigData, and COVID-19. For single-label, multi-label, and fine-grained classification, our model achieves state-of-the-art performance, the classification accuracy has increased by a maximum of 6.69\%. The code is available in https://github.com/PerceptionComputingLab/MedFILIP.

cs.CV↗

arXiv2024

PointCFormer: a Relation-based Progressive Feature Extraction Network for Point Cloud Completion

Point cloud completion aims to reconstruct the complete 3D shape from incomplete point clouds, and it is crucial for tasks such as 3D object detection and segmentation. Despite the continuous advances in point cloud analysis techniques, feature extraction methods are still confronted with apparent limitations. The sparse sampling of point clouds, used as inputs in most methods, often results in a certain loss of global structure information. Meanwhile, traditional local feature extraction methods usually struggle to capture the intricate geometric details. To overcome these drawbacks, we introduce PointCFormer, a transformer framework optimized for robust global retention and precise local detail capture in point cloud completion. This framework embraces several key advantages. First, we propose a relation-based local feature extraction method to perceive local delicate geometry characteristics. This approach establishes a fine-grained relationship metric between the target point and its k-nearest neighbors, quantifying each neighboring point's contribution to the target point's local features. Secondly, we introduce a progressive feature extractor that integrates our local feature perception method with self-attention. Starting with a denser sampling of points as input, it iteratively queries long-distance global dependencies and local neighborhood relationships. This extractor maintains enhanced global structure and refined local details, without generating substantial computational overhead. Additionally, we develop a correction module after generating point proxies in the latent space to reintroduce denser information from the input points, enhancing the representation capability of the point proxies. PointCFormer demonstrates state-of-the-art performance on several widely used benchmarks. Our code is available at https://github.com/Zyyyyy0926/PointCFormer_Plus_Pytorch.

cs.CV↗

arXiv2024

PSpectCosmo: A Pseudo-Spectral Code for Cosmological Dynamics Spanning Inflation and Reheating

\texttt{PSpectCosmo} is a high-performance \texttt{C++} program developed to investigate early-universe cosmological dynamics, with a specific emphasis on the inflationary epoch. Utilizing a Fourier-space pseudo-spectral method, \texttt{PSpectCosmo} enables the precise evolution of interacting scalar fields and gravitational waves, ensuring accurate representation of the power spectrum during inflation. This approach overcomes key limitations of finite difference methods, particularly in maintaining consistency between effective and lattice wave vectors. The code employs the adaptive step size velocity-Verlet algorithm for time integration, offering a balance of numerical stability and high precision. Additionally, \texttt{PSpectCosmo} incorporates a robust mechanism to compute convergent energy density, effectively resolving the issue of divergent energy density at the onset of inflation. These capabilities establish \texttt{PSpectCosmo} as a reliable and versatile tool for probing non-linear cosmological phenomena and inflationary dynamics with exceptional accuracy. The code is publicly available at https://github.com/JieJiang-Cosmology/PSpectCosmo.git

astro-ph.CO↗

arXiv2024

Global boundedness induced by asymptotically non-degenerate motility in a fully parabolic chemotaxis model with local sensing

A fully parabolic chemotaxis model of Keller-Segel type with local sensing is considered. The system features a signal-dependent asymptotically non-degenerate motility function, which accounts for a repulsion-dominated chemotaxis. Global boundedness of classical solutions is proved for an initial Neumann boundary value problem of the system in any space dimension. In addition, stabilization towards the homogeneous steady state is shown, provided that the motility is monotone non-decreasing. The key steps of the proof consist of the introduction of several auxiliary functions and a refined comparison argument, along with uniform-in-time estimates for a functional involving nonlinear coupling between the unknowns and auxiliary functions.

math.AP↗

arXiv2024

Scaling Mesh Generation via Compressive Tokenization

We propose a compressive yet effective mesh representation, Blocked and Patchified Tokenization (BPT), facilitating the generation of meshes exceeding 8k faces. BPT compresses mesh sequences by employing block-wise indexing and patch aggregation, reducing their length by approximately 75\% compared to the original sequences. This compression milestone unlocks the potential to utilize mesh data with significantly more faces, thereby enhancing detail richness and improving generation robustness. Empowered with the BPT, we have built a foundation mesh generative model training on scaled mesh data to support flexible control for point clouds and images. Our model demonstrates the capability to generate meshes with intricate details and accurate topology, achieving SoTA performance on mesh generation and reaching the level for direct product usage.

cs.GR↗

arXiv2024

Manipulating ferroelectric topological polar structures with twisted light

The dynamic control of novel states of matter beyond thermodynamic equilibrium is a fundamental pursuit in condensed matter physics. Intense terahertz fields have enabled metal-insulator transitions, superconductivity, quantum paraelectric ferroelectricity, and room-temperature magnetization via circularly polarized terahertz electric fields. These effects hinge on the excitation of infrared-active soft phonon modes by terahertz fields. Expanding this concept, recent theory suggests that ferroelectric polarization may be manipulated through terahertz twisted light, transferring orbital angular momentum to create ferroelectric skyrmions. Our study experimentally demonstrates that such control is possible in quasi-2D ferroelectric CsBiNb2O7 using twisted UV light with orbital angular momentum (OAM). By resonantly exciting both the ferroelectric mode and the octahedral tilting mode, twisted UV light dynamically modulates the ferroelectric polarization. We employ in-situ X-ray Bragg coherent diffractive imaging, twisted optical Raman spectroscopy, and density functional theory to three-dimensionally resolve ionic displacement fields and polarization texture changes. Our observations reveal deterministic, reversible twisted light-induced strain and ionic displacements within the unit cell, causing substantial microscopic polarization changes. This interaction between twisted photons, phonon modes, and induced ionic displacements breaks symmetry and stabilizes a non-equilibrium ferroelectric phase with topological solitons. These findings offer a new path to control ferroelectricity and magnetism, opening avenues for novel optoelectronic devices such as ultrafast non-volatile memory switches by using light to coherently control ferroic states.

cond-mat.mtrl-sci↗

arXiv2024

Parametric resonance of gravitational waves in general scalar-tensor theories

Gravitational waves offer a potent mean to test the underlying theory of gravity. In general theories of gravity, such as scalar-tensor theories, one expects modifications in the friction term and the sound speed in the gravitational wave equation. In that case, rapid oscillations in such coefficients, e.g. due to an oscillating scalar field, may lead to narrow parametric resonances in the gravitational wave strain. We perform a general analysis of such possibility within DHOST theories. We use disformal transformations to find the theory space with larger resonances, within an effective field theory approach. We then apply our formalism to a non-minimally coupled ultra-light dark matter scalar field, assuming the presence of a primordial gravitational wave background, e.g., from inflation. We find that the resonant peaks in the spectral density may be detectable by forthcoming detectors such as LISA, Taiji, Einstein Telescope and Cosmic Explorer.

gr-qc↗

arXiv2024

Towards Scalable Semantic Representation for Recommendation

With recent advances in large language models (LLMs), there has been emerging numbers of research in developing Semantic IDs based on LLMs to enhance the performance of recommendation systems. However, the dimension of these embeddings needs to match that of the ID embedding in recommendation, which is usually much smaller than the original length. Such dimension compression results in inevitable losses in discriminability and dimension robustness of the LLM embeddings, which motivates us to scale up the semantic representation. In this paper, we propose Mixture-of-Codes, which first constructs multiple independent codebooks for LLM representation in the indexing stage, and then utilizes the Semantic Representation along with a fusion module for the downstream recommendation stage. Extensive analysis and experiments demonstrate that our method achieves superior discriminability and dimension robustness scalability, leading to the best scale-up performance in recommendations.

cs.IR↗

arXiv2024

Self-Bootstrapped Visual-Language Model for Knowledge Selection and Question Answering

While large visual-language models (LVLM) have shown promising results on traditional visual question answering benchmarks, it is still challenging for them to answer complex VQA problems which requires diverse world knowledge. Motivated by the research of retrieval-augmented generation in the field of natural language processing, we use Dense Passage Retrieval (DPR) to retrieve related knowledge to help the model answer questions. However, DPR conduct retrieving in natural language space, which may not ensure comprehensive acquisition of image information. Thus, the retrieved knowledge is not truly conducive to helping answer the question, affecting the performance of the overall system. To address this issue, we propose a novel framework that leverages the visual-language model to select the key knowledge retrieved by DPR and answer questions. The framework consists of two modules: Selector and Answerer, where both are initialized by the LVLM and parameter-efficiently finetuned by self-bootstrapping: find key knowledge in the retrieved knowledge documents using the Selector, and then use them to finetune the Answerer to predict answers; obtain the pseudo-labels of key knowledge documents based on the predictions of the Answerer and weak supervision labels, and then finetune the Selector to select key knowledge; repeat. Our framework significantly enhances the performance of the baseline on the challenging open-domain Knowledge-based VQA benchmark, OK-VQA, achieving a state-of-the-art accuracy of 62.83%. Our code is publicly available at https://github.com/haodongze/Self-KSel-QAns.

cs.CV↗

arXiv2024

Deep learning for cosmological parameter inference from a dark matter halo density field

We propose a lightweight deep convolutional neural network (lCNN) to estimate cosmological parameters from simulated three-dimensional dark matter (DM) halo distributions and associated statistics. The training dataset comprises 2000 realizations of a cubic box with a side length of 1000 $h^{-1}{\rm Mpc}$, and interpolated over a cubic grid of $300^3$ voxels, with each simulation produced using $512^3$ DM particles and $512^3$ neutrinos. Under the flat $Λ$CDM model, simulations vary standard six cosmological parameters including $Ω_m$, $Ω_b$, $h$, $n_s$, $σ_8$, $w$, along with the neutrino mass sum, $M_ν$. We find that: 1) within the framework of lCNN, extracting large-scale structure information is more efficient from the halo density field compared to relying on the statistical quantities including the power spectrum, the two-point correlation function, and the coefficients from wavelet scattering transform; 2) combining the halo density field with its Fourier transformed counterpart enhances predictions, while augmenting the training dataset with measured statistics further improves performance; 3) achieving high accuracy in inferring $Ω_m$, $h$, and $σ_8$ by the neural network model, while being inefficient in predicting $Ω_b$, { $n_s$}, $M_ν$ and $w$; 4) { compared to the simple fully connected network trained with three statistical quantities, our CNN yields statistically reduced errors, showing improvements of approximately 23\% for $Ω_m$, 11\% for $h$, 8\% for $n_s$, and 21\% for $σ_8$. Additionally, in comparison with the likelihood-based analysis on $P(k)$ data, our CNN provides much tighter constraints on parameters, especially on $Ω_m$ and $σ_8$.} Our study emphasizes this lCNN-based novel approach in extracting large-scale structure information and estimating cosmological parameters.

astro-ph.CO↗

arXiv2024

Long-term variation of the solar polar magnetic fields at different latitudes

The polar magnetic fields of the Sun play an important role in governing solar activity and powering fast solar wind. However, because our view of the Sun is limited in the ecliptic plane, the polar regions remain largely uncharted. Using the high spatial resolution and polarimetric precision vector magnetograms observed by Hinode from 2012 to 2021, we investigate the long-term variation of the magnetic fields in polar caps at different latitudes. The Hinode magnetic measurements show that the polarity reversal processes in the north and south polar caps are non-simultaneous. The variation of the averaged radial magnetic flux density reveals that, in each polar cap, the polarity reversal is completed successively from the 70 degree latitude to the pole, reflecting a poleward magnetic flux migration therein. These results clarify the polar magnetic polarity reversal process at different latitudes.

astro-ph.SR↗

arXiv2024

Meridional flow in the solar polar caps revealed by magnetic field observation and simulation

As a large-scale motion on the Sun, the meridional flow plays an important role in determining magnetic structure and strength and solar cycle. However, the meridional flow near the solar poles is still unclear. The Hinode observations show that the magnetic flux density in polar caps decreases from the lower latitudes to the poles. Using a surface flux transport model, we simulate the global radial magnetic field to explore the physical process leading to the observed polar magnetic distribution pattern. For the first time, the high-resolution observations of the polar magnetic fields observed by Hinode are used to directly constrain the simulation. Our simulation reproduces the observed properties of the polar magnetic fields, suggesting the existence of a counter-cell meridional flow in the solar polar caps with a maximum amplitude of about 3 m s$^{-1}$.

astro-ph.SR↗

arXiv2024

Understanding the Ranking Loss for Recommendation with Sparse User Feedback

Click-through rate (CTR) prediction is a crucial area of research in online advertising. While binary cross entropy (BCE) has been widely used as the optimization objective for treating CTR prediction as a binary classification problem, recent advancements have shown that combining BCE loss with an auxiliary ranking loss can significantly improve performance. However, the full effectiveness of this combination loss is not yet fully understood. In this paper, we uncover a new challenge associated with the BCE loss in scenarios where positive feedback is sparse: the issue of gradient vanishing for negative samples. We introduce a novel perspective on the effectiveness of the auxiliary ranking loss in CTR prediction: it generates larger gradients on negative samples, thereby mitigating the optimization difficulties when using the BCE loss only and resulting in improved classification ability. To validate our perspective, we conduct theoretical analysis and extensive empirical evaluations on public datasets. Additionally, we successfully integrate the ranking loss into Tencent's online advertising system, achieving notable lifts of 0.70% and 1.26% in Gross Merchandise Value (GMV) for two main scenarios. The code is openly accessible at: https://github.com/SkylerLinn/Understanding-the-Ranking-Loss.

cs.IR↗