arXiv subjects

Yangrui Chen

Publications and source records attributed to Yangrui Chen.

At least 19 recordsLinked to original sources

arXiv2026

Statistical Study of Solar Prominence Plumes Based on NVST H$\alpha$ Observations

Plumes are one of the most representative dynamic features observed in prominences and play a key role in mass and magnetic transport within them. However, their physical nature and triggering processes remain actively debated. Based on limb H$\alpha$ observations from the New Vacuum Solar Telescope (NVST) during 2013--2025, we statistically investigated 34 plumes with clear and complete evolutions by developing an automated image-processing pipeline. It is revealed that plume lifetimes mainly range from 300 s to 700 s, with vertical displacements between 3--7 Mm. The mean widths and velocities are concentrated in the range of 0.5--1.5 Mm and 10--20 km s$^{-1}$, respectively. Besides wide distribution ranges, plume parameters exhibit irregular evolution fluctuations, indicating that the formation and evolution of various plumes may exhibit different physical patterns. Correlation analysis among the parameters further reveals that: (1) Positive correlations were found among lifetime, vertical displacement, and mean width, indicating an intrinsic coupling between the temporal and spatial scales of plumes. (2) Trajectory curvature is negatively correlated with lifetime, vertical displacement, and velocity. Accelerating and width-contracting plumes typically have lower curvature, suggesting that curvature may reflect environmental influences and the stability of plumes. (3) Plumes with higher initial velocities were more likely to be accompanied by precursor brightening, suggesting that these plumes may be triggered by magnetic reconnection. Furthermore, we infer that some plumes in non-bubble regions may be inherently driven by mini-filament eruptions. These results establish a statistical framework for prominence plumes and reveal diversity in their dynamical evolution and triggering mechanisms.

astro-ph.SR↗

arXiv2026

Can We Distinguish the Source Region Location of Filament/Prominence Eruptions from the Sun-as-a-star H$\alpha$ Spectrum?

Solar filament/prominence eruptions can significantly perturb geospace when originating from favorable source locations and directions. While stellar analogs have been recently reported, the disk locations and magnetic environments of their source regions remain spatially unresolved on other stars. To bridge this gap, we investigate the typical Sun-as-a-star H$\alpha$ temporal spectral characteristics of solar filament/prominence eruptions with different source region locations (on-disk vs. limb, active region vs. quiet-Sun region). It is revealed that limb eruptions are characterized by blueshifted/redshifted emission caused by the bright off-limb erupting structures, whereas on-disk eruptions may show blueshifted absorptions due to the dark erupting filaments. Among the limb eruptions, front-side limb eruptions usually display line center emission before the blueshifted/redshifted emission, while far-side limb eruptions show the opposite sequence. Moreover, the magnetic environment at source also shapes the spectral characteristics. On-disk filament eruptions from active region exhibit much more intense flare-ribbon-dominated line center emission features compared with those from quiet-Sun region. Limb active region eruptions often show single-wing emissions, whereas large-scale quiet-Sun region (quiescent) prominence eruptions frequently display expansion-induced emission in both wings followed by line center absorption due to the disappearance of bright prominence. These distinct Sun-as-a-star H$\alpha$ spectral characteristics, dependent on eruption location, provide a diagnostic basis for inferring source regions of stellar filament/prominence eruptions from spatially unresolved H$\alpha$ spectra.

astro-ph.SR↗

arXiv2026

MegaScale-Omni: A Hyper-Scale, Workload-Resilient System for MultiModal LLM Training in Production

As the foundational component of versatile AI applications, training an multimodal large language model (MLLM) relies on multimodal datasets with dynamic modality mixture proportions and sample length distributions. However, existing MLLM systems remain inefficient under dynamic workloads, due to statically coupled decisions of resource allocation and model parallelization between encoders and the LLM backbone. This paper presents MegaScale-Omni, an industrial-grade MLLM training system tailored for dynamic workload adaption and hyper-scale deployment. MegaScale-Omni is built upon the training scheme of encoder-LLM multiplexing with three key innovations: (1) Decoupled parallelism strategies with long-short sequence parallelism for encoders to process variable-length samples, and full-fledged 5D parallelism for the LLM backbone, both organized under a communication-efficient parallelization layout. (2) Unified encoder-LLM representations for flexible, extensible colocation, and a new paradigm of encoder-LLM joint pipeline with workload resilience. (3) Workload balancing techniques via decentralized grouped reordering in data loaders and adaptive resharding from encoder to LLM ranks. MegaScale-Omni is deployed as the foundation of our in-house large-scale MLLM training tasks with thousands of GPUs. Our experimental results demonstrate $1.27\times$-$7.57\times$ throughput improvement under production-grade dynamic workloads, as compared to four state-of-the-art systems.

cs.DC↗

arXiv2025

Virtual Width Networks

We introduce Virtual Width Networks (VWN), a framework that delivers the benefits of wider representations without incurring the quadratic cost of increasing the hidden size. VWN decouples representational width from backbone width, expanding the embedding space while keeping backbone compute nearly constant. In our large-scale experiment, an 8-times expansion accelerates optimization by over 2 times for next-token and 3 times for next-2-token prediction. The advantage amplifies over training as both the loss gap grows and the convergence-speedup ratio increases, showing that VWN is not only token-efficient but also increasingly effective with scale. Moreover, we identify an approximately log-linear scaling relation between virtual width and loss reduction, offering an initial empirical basis and motivation for exploring virtual-width scaling as a new dimension of large-model efficiency.

cs.LG↗

arXiv2025

Mapping ground-based coronagraphic images to Helioprojective-Cartesian coordinate system by image registration

A few ground-based solar coronagraphs have been installed in western China for observing the low-layer corona in recent years. However, determining the Helioprojective Coordinates for the coronagraphic data with high precision is an important but challenging step for further research with other multi-wavelength data. In this paper, we propose an automatic coronal image registration method that combines local statistical correlation and feature point matching to achieve accurate registration between ground-based coronal green-line images and space-based 211 {\AA} images. Then, the accurate field of view information of the coronal green-line images can be derived, allowing the images to be mapped to the Helioprojective Cartesian Coordinates with an accuracy of no less than 0.1''. This method has been extensively validated using 100 days of coronal data spanning an 11-year period, demonstrating its broad applicability to ground-based coronagraphs equipped with green-line observations. It significantly enhances the scientific value of ground-based coronal data, enabling comprehensive studies of coronal transient activities and facilitating the joint analysis of data from multiple instruments. Additionally, it holds potential for future applications in improving the pointing accuracy of coronagraphs.

astro-ph.SR↗

arXiv2025

Seed1.5-VL Technical Report

We present Seed1.5-VL, a vision-language foundation model designed to advance general-purpose multimodal understanding and reasoning. Seed1.5-VL is composed with a 532M-parameter vision encoder and a Mixture-of-Experts (MoE) LLM of 20B active parameters. Despite its relatively compact architecture, it delivers strong performance across a wide spectrum of public VLM benchmarks and internal evaluation suites, achieving the state-of-the-art performance on 38 out of 60 public benchmarks. Moreover, in agent-centric tasks such as GUI control and gameplay, Seed1.5-VL outperforms leading multimodal systems, including OpenAI CUA and Claude 3.7. Beyond visual and video understanding, it also demonstrates strong reasoning abilities, making it particularly effective for multimodal reasoning challenges such as visual puzzles. We believe these capabilities will empower broader applications across diverse tasks. In this report, we mainly provide a comprehensive review of our experiences in building Seed1.5-VL across model design, data construction, and training at various stages, hoping that this report can inspire further research. Seed1.5-VL is now accessible at https://www.volcengine.com/ (Volcano Engine Model ID: doubao-1-5-thinking-vision-pro-250428)

cs.CV↗

arXiv2025

MegaScale-Data: Scaling Dataloader for Multisource Large Foundation Model Training

Modern frameworks for training large foundation models (LFMs) employ dataloaders in a data-parallel manner, with each loader processing a disjoint subset of training data. When preparing data for LFM training that originates from multiple, distinct sources, two fundamental challenges arise. First, due to the quadratic computational complexity of the attention operator, the non-uniform sample distribution over data-parallel ranks leads to significant workload imbalance among dataloaders, degrading the training efficiency. Second, supporting diverse data sources requires per-dataset file access states that are redundantly replicated across parallel loaders, consuming excessive memory. This also hinders dynamic data mixing (e.g., curriculum learning) and causes redundant access/memory overhead in hybrid parallelism. We present MegaScale-Data, an industrial-grade distributed data loading architecture for multisource LFMs training, with three key innovations: (1) Disaggregated data preprocessing via role-specific actors (Source Loaders/Data Constructors) to eliminate source and parallelism redundant data access and ensure multisource scalability. (2) Centralized and declarative data plane for load-time multisource orchestration, such as long-short context, multimodality, and curriculum learning. (3) Multi-level auto-partitioning and scaling mechanism for source loaders under heterogeneous preprocessing costs. We also contribute our designs and operational experience in deployment and fault tolerance. MegaScale-Data achieves up to: (1) 4.5x end-to-end training throughput improvement, and (2) 13.5x reduction in CPU memory usage.

cs.DC↗

arXiv2025

Topological signatures of collective dynamics and turbulent-like energy cascades in apolar active granular

Active matter refers to a broad class of non-equilibrium systems where energy is continuously injected at the level of individual ``particles". These systems exhibit emergent collective behaviors that have no direct thermal-equilibrium counterpart. Their scale ranges from micrometer-sized swarms of bacteria to meter-scale human crowds. In recent years, the role of topology and self-propelled topological defects in active systems has garnered significant attention, particularly in polar and nematic active matter. Building on these ideas, we investigate emergent collective dynamics in apolar active granular fluids. Using isotropic granular vibrators as a model experimental system of apolar active Ornstein-Uhlenbeck particles in a dry environment, we uncover a distinctive three-stage time evolution arising from the intricate interplay between activity and inelastic interactions. By analyzing the statistics, spatial correlations, and dynamics of vortex-like topological defects in the displacement vector field, we demonstrate their ability to describe this intrinsic collective motion. Furthermore, associated to these topological defects, we reveal the onset of a turbulent-like inverse energy cascade, where kinetic energy transfers across different length scales over time. As the system evolves, the power scaling of the energy transfer increases with the duration of observation. Our findings show that topological concepts can be extended to the nonequilibrium dynamics of apolar active matter, revealing a direct link between microscopic topological processes and emergent large-scale behaviors in active granular fluids that lack both a well-defined direction of motion and an intrinsic axis of orientation at the particle scale.

cond-mat.soft↗

arXiv2024

Optimus: Accelerating Large-Scale Multi-Modal LLM Training by Bubble Exploitation

Multimodal large language models (MLLMs) have extended the success of large language models (LLMs) to multiple data types, such as image, text and audio, achieving significant performance in various domains, including multimodal translation, visual question answering and content generation. Nonetheless, existing systems are inefficient to train MLLMs due to substantial GPU bubbles caused by the heterogeneous modality models and complex data dependencies in 3D parallelism. This paper proposes Optimus, a distributed MLLM training system that reduces end-to-end MLLM training time. Optimus is based on our principled analysis that scheduling the encoder computation within the LLM bubbles can reduce bubbles in MLLM training. To make scheduling encoder computation possible for all GPUs, Optimus searches the separate parallel plans for encoder and LLM, and adopts a bubble scheduling algorithm to enable exploiting LLM bubbles without breaking the original data dependencies in the MLLM model architecture. We further decompose encoder layer computation into a series of kernels, and analyze the common bubble pattern of 3D parallelism to carefully optimize the sub-millisecond bubble scheduling, minimizing the overall training time. Our experiments in a production cluster show that Optimus accelerates MLLM training by 20.5%-21.3% with ViT-22B and GPT-175B model over 3072 GPUs compared to baselines.

cs.CL↗

arXiv2024

Dispersionless Flat Mode and Vibrational Anomaly in Active Brownian Vibrators Induced by String-like Dynamical Defects

In recent years, active Brownian particles have emerged as a prominent model system for comprehending the behaviors of active matter, wherein particles demonstrate self-propelled motion by harnessing energy from the surrounding environment. A fundamental objective of studying active matter is to elucidate the physical mechanisms underlying its collective behaviors. Drawing inspiration from advancements in molecular glasses, our study unveils a low-energy ``flat mode" within the transverse spectrum of active Brownian vibrators -- a nearly two-dimensional, bi-disperse granular assembly. We demonstrate that this collective excitation induces an anomalous excess in the vibrational density of states (VDOS) beyond the phononic Debye contribution. We characterize the properties of this flat mode by exploring the parameter space of our experimental system and tuning the packing fraction, the vibrational frequency, the particle size ratio and the mixture ratio. Additionally, we establish through empirical evidence that string-like dynamical defects, discerned via the spatial distribution of each particle's contribution to the reduced transverse VDOS, serve as the microscopic origin of the flat mode and its associated anomalies.

cond-mat.soft↗

arXiv2024

Experimental observation of gapped shear waves and liquid-like to gas-like dynamical crossover in active granular matter

Unlike crystalline solids, liquids lack long-range order, resulting in diffusive shear fluctuations rather than propagating waves. Simulations predict that liquids exhibit a $k$-gap in wave-vector space, where solid-like transverse waves reappear above this gap. Experimental evidence in classical liquids has been limited, observed only in 2D dusty plasmas. Here, we investigate this phenomenon using active Brownian vibrators and uncover distinct gas-like and liquid-like phases depending on the packing fraction. We measure key properties, including pair correlation functions, mean square displacements, velocity auto-correlation functions, and vibrational density of states. In the liquid-like phase, we confirm the $k$-gap in transverse excitations, whose size grows as the packing fraction decreases and eventually disappears in the gas phase. Our findings extend the concept of the $k$-gap to active granular systems and reveal striking parallels with supercritical fluids.

cond-mat.soft↗

arXiv2024

MegaScale: Scaling Large Language Model Training to More Than 10,000 GPUs

We present the design, implementation and engineering experience in building and deploying MegaScale, a production system for training large language models (LLMs) at the scale of more than 10,000 GPUs. Training LLMs at this scale brings unprecedented challenges to training efficiency and stability. We take a full-stack approach that co-designs the algorithmic and system components across model block and optimizer design, computation and communication overlapping, operator optimization, data pipeline, and network performance tuning. Maintaining high efficiency throughout the training process (i.e., stability) is an important consideration in production given the long extent of LLM training jobs. Many hard stability issues only emerge at large scale, and in-depth observability is the key to address them. We develop a set of diagnosis tools to monitor system components and events deep in the stack, identify root causes, and derive effective techniques to achieve fault tolerance and mitigate stragglers. MegaScale achieves 55.2% Model FLOPs Utilization (MFU) when training a 175B LLM model on 12,288 GPUs, improving the MFU by 1.34x compared to Megatron-LM. We share our operational experience in identifying and fixing failures and stragglers. We hope by articulating the problems and sharing our experience from a systems perspective, this work can inspire future LLM systems research.

cs.LG↗

arXiv2023

Anomalous large-scale collective motion in granular Brownian vibrators

Using Brownian vibrators, we conducted a study on the structures and dynamics of quasi-2d granular materials with packing fractions ($ϕ$) ranging from 0.111 to 0.832. Our observations revealed a remarkable large-scale collective motion in hard granular disk systems, encompassing four distinct phases: granular fluid, collective fluid, poly-crystal, and crystal. The collective motion emerge at $ϕ=$0.317, coinciding with a peak in local density fluctuations. However, this collective motion ceased to exist at $ϕ=$0.713 when the system transitioned into a crystalline state. While the poly-crystal and crystal phases exhibited similarities to equilibrium hard disks, the first two phases differed significantly from the equilibrium systems and previous experiments involving uniformly driven spheres. This disparity suggests that the collective motion arises from a competition controlled by volume fraction. Specifically, it involves an active force and an effective attractive interaction resulting from inelastic particle collisions. Remarkably, these findings align with recent theoretical research on the flocking motion of spherical active particles without alignment mechanisms.

cond-mat.soft↗

arXiv2023

Clusters and collective motions in Brownian vibrators

Using Brownian vibrators, where single particles can undergo Brownian motion under vibration, we experimentally investigated self-organized structures and dynamics of quasi-two-dimensional (quasi-2d) granular materials with volume fractions $0.111\leϕ\le0.832$. We show rich structures and dynamics in hard-disk systems of inelastic particle collisions, with four phases corresponding to cluster fluid, collective fluid, poly-crystal, and crystal. While poly-crystal and crystal are strikingly similar to the equilibrium hard disks, the first two phases differ substantially from the equilibrium ones and the previous quasi-2d experiments of uniformly driven spheres. Our investigation provides single-particle-scale evidence that granular materials subject to uniform random forcing are weakly cohesive with complex internal structures and dynamics. Moreover, our experiment shows that large-scale collective motion can arise in a purely repulsive hard-disk system. The collective motion emerges near $ϕ=0.317$, where the most significant clusters span half of the system, and disappears near $ϕ=0.713$, around which the system crystallizes and the melting transition occurs in the equilibrium hard disks.

cond-mat.soft↗

arXiv2022

High-energy velocity tails in uniformly heated granular materials

We experimentally investigate the velocity distributions of quasi two-dimensional granular materials, which are homogeneously driven, i.e. uniformly heated, by an electromagnetic vibrator, where the translational velocity and the rotation of a single particle are Gaussian and independent. We observe the non-Gaussian distributions of particle velocity, with the density-independent high-energy tails characterized by an exponent of $β=1.50\pm0.03$ for volume fractions of $0.111\leϕ\le0.832$, covering a wide range of structures and dynamics. Surprisingly, our results are in excellent agreement with the prediction of the kinetic theories of granular gas, even for an extremely high volume fraction of $ϕ=0.832$ where the granular material forms a crystalline solid. Our experiment reveals that the density-independent high-energy velocity tails of $β=1.50$ are a fundamental property of uniformly heated granular matter.

cond-mat.soft↗

arXiv2021

BGL: GPU-Efficient GNN Training by Optimizing Graph Data I/O and Preprocessing

Graph neural networks (GNNs) have extended the success of deep neural networks (DNNs) to non-Euclidean graph data, achieving ground-breaking performance on various tasks such as node classification and graph property prediction. Nonetheless, existing systems are inefficient to train large graphs with billions of nodes and edges with GPUs. The main bottlenecks are the process of preparing data for GPUs - subgraph sampling and feature retrieving. This paper proposes BGL, a distributed GNN training system designed to address the bottlenecks with a few key ideas. First, we propose a dynamic cache engine to minimize feature retrieving traffic. By a co-design of caching policy and the order of sampling, we find a sweet spot of low overhead and high cache hit ratio. Second, we improve the graph partition algorithm to reduce cross-partition communication during subgraph sampling. Finally, careful resource isolation reduces contention between different data preprocessing stages. Extensive experiments on various GNN models and large graph datasets show that BGL significantly outperforms existing GNN training systems by 20.68x on average.

cs.LG↗

arXiv2021

Critical behaviors of jamming in cyclically sheared frictional hard granular particles

In stark contrast to the jamming of frictionless hard-sphere (hard-disk if in two dimensions) packings, the critical jamming of frictional packings is much more elusive and still under intense debate. Here we show that frictional hard-disk packings self-organize near a critical jamming state when subjected to quasi-static steady-state cyclic shear, displaying scale-free fluctuations in particle velocity fields as characterized by the power spectra $E(k)\propto k^{-α}$ for both longitudinal and transverse modes in wavevector space, with $α\approx2.0\pm0.15$, and the nearly flat spectra $E(ω)\propto const.$ in angular frequency domain. Our findings agree quantitatively with the predictions of the Langevin-type effective medium theory of S. Henkes and coworkers with two diverging length scales associated respectively with the longitudinal and transverse modes, showing a hallmark of critical behaviors of jamming. Moreover, our findings are consistent with the conceptual framework of general isostaticity but with a key difference that the system self-organizes to a critical jamming state through a strong coupling of mechanical structure and dynamics. Our findings are important in providing microscopic mechanism in understanding the nonlocal rheologies and guide principles in developing constitutive relations of real granular materials.

cond-mat.soft↗

arXiv2021

Disorder-induced vibrational anomalies from crystalline to amorphous solids

The origin of boson peak -- an excess of density of states over Debye's model in glassy solids -- is still under intense debate, among which some theories and experiments suggest that boson peak is related to van-Hove singularity. Here we show that boson peak and van-Hove singularity are well separated identities, by measuring the vibrational density of states of a two-dimensional granular system, where packings are tuned gradually from a crystalline, to polycrystals, and to an amorphous material. We observe a coexistence of well separated boson peak and van-Hove singularities in polycrystals, in which the van-Hove singularities gradually shift to higher frequency values while broadening their shapes and eventually disappear completely when the structural disorder $η$ becomes sufficiently high. By analyzing firstly the strongly disordered system ($η=1$) and the disordered granular crystals ($η=0$), and then systems of intermediate disorder with $η$ in between, we find that boson peak is associated with spatially uncorrelated random flucutations of shear modulus $δG/\langle G \rangle$ whereas the smearing of van-Hove singularities is associated with spatially correlated fluctuations of shear modulus $δG/\langle G \rangle$.

cond-mat.soft↗