arXiv subjects

Dong He

Publications and source records attributed to Dong He.

At least 19 recordsLinked to original sources

arXiv2026

ScaleAcross Explorer: Exploring Communication Optimization for Scale-Across AI Model Training

The rapid scaling of large language model training requires distributing GPU resources across multiple data center buildings and regions. We refer to such paradigm as "scale-across" training. As infrastructure expands, the system design space becomes increasingly intricate, encompassing new model architectures, hardware heterogeneity, and evolving communication patterns. Drawing from Meta's production experience, we highlight the complexities of deploying training jobs across a few data centers housing hundreds of thousands of GPUs. To accelerate exploration of the large design space and to enable efficient training for frontier model development, we conduct in-depth characterization of three key design dimensions: parallelism placement, parallelism scheduling, and network layer technologies. We then propose ScaleAcross Explorer, an optimizer that considers the interplay of design dimensions and holistically optimizes scale-across training. Testbed experiments and simulations demonstrate up to 64.62% training speedups over production configuration and up to 37.59% training speedups over the state-of-the-art baseline across a wide range of design points.

cs.DC↗

arXiv2025

On the Ding and Helleseth's 9th open problem about optimal ternary cyclic codes

The cyclic code is a subclass of linear codes and has applications in consumer electronics, data storage systems and communication systems as they have efficient encoding and decoding algorithms. In 2013, Ding, et al. presented nine open problems about optimal ternary cyclic codes. Till now, the 1st, 2nd and 6th problems were completely solved, and the 3rd, 7th, 8th and 9th problems were partially solved. In this manuscript, we focus on the 9th problem. By determining the root set of some special polynomials over finite fields, we give an incomplete answer for the 9th problem, and then we construct two classes of optimal ternary cyclic codes with respect to the Sphere Packing Bound basing on some special polynomials over finite fields

cs.IT↗

arXiv2025

On the Ding and Helleseth's 8th open problem about optimal ternary cyclic codes

The cyclic code is a subclass of linear codes and has applications in consumer electronics, data storage systems and communication systems due to the efficient encoding and decoding algorithms. In 2013, Ding, et al. presented nine open problems about optimal ternary cyclic codes. Till now, the 1st, 2nd, 6th and 7th problems were completely solved, the 3rd, 8th and 9th problems were incompletely solved. In this manuscript, we focus on the 8th problem. By determining the root set of some special polynomials over finite fields, we present a counterexample and a sufficient condition for the ternary cyclic code $\mathcal{C}_{(1, e)}$ optimal. Furthermore, basing on the properties of finite fields, we construct a class of optimal ternary cyclic codes with respect to the Sphere Packing Bound, and show that these codes are not equivalent to any known codes.

cs.IT↗

arXiv2025

The error-correcting pair for several classes of NMDS linear codes

The error-correcting pair is a general algebraic decoding method for linear codes. The near maximal distance separable (NMDS) linear code is a subclass of linear codes and has applications in secret sharing scheme and communication systems due to the efficient performance, thus we focus on the error-correcting pair of NMDS linear codes. In 2023, He and Liao showed that for an NMDS linear code $\mathcal{C}$ with minimal distance $2\ell+1$ or $2\ell+2$, if $\mathcal{C}$ has an $\ell$-error-correcting pair $\left( \mathcal{A}, \mathcal{B} \right)$, then the parameters of $\mathcal{A}$ have 6 or 10 possibilities, respectively. In this manuscript, basing on Product Singleton Bound, we give several necessary conditions for that the NMDS linear code $\mathcal{C}$ with minimal distance $2\ell+1$ has an $\ell$-error-correcting pair $(\mathcal{A}, \mathcal{B})$, where the parameters of $\mathcal{A}$ is the 1st, 2nd, 4th or 5th case, then basing on twisted generalized Reed-Solomon codes, we give an example for that the parameters of $\mathcal{A}$ is the 1st case. Moreover, we also give several necessary conditions for that the NMDS linear code $\mathcal{C}$ with minimal distance $2\ell+2$ has an $\ell$-error-correcting pair $(\mathcal{A}, \mathcal{B})$, where the parameters of $\mathcal{A}$ is the 2nd, 4th, 7th or 8th case, then we give an example for that the parameters of $\mathcal{A}$ is the 1st or 2nd case, respectively.

cs.IT↗

arXiv2024

Behavior evolution-inspired approach to walking gait reinforcement training for quadruped robots

Reinforcement learning method is extremely competitive in gait generation techniques for quadrupedal robot, which is mainly due to the fact that stochastic exploration in reinforcement training is beneficial to achieve an autonomous gait. Nevertheless, although incremental reinforcement learning is employed to improve training success and movement smoothness by relying on the continuity inherent during limb movements, challenges remain in adapting gait policy to diverse terrain and external disturbance. Inspired by the association between reinforcement learning and the evolution of animal motion behavior, a self-improvement mechanism for reference gait is introduced in this paper to enable incremental learning of action and self-improvement of reference action together to imitate the evolution of animal motion behavior. Further, a new framework for reinforcement training of quadruped gait is proposed. In this framework, genetic algorithm is specifically adopted to perform global probabilistic search for the initial value of the arbitrary foot trajectory to update the reference trajectory with better fitness. Subsequently, the improved reference gait is used for incremental reinforcement learning of gait. The above process is repeatedly and alternatively executed to finally train the gait policy. The analysis considering terrain, model dimensions, and locomotion condition is presented in detail based on simulation, and the results show that the framework is significantly more adaptive to terrain compared to regular incremental reinforcement learning.

cs.RO↗

arXiv2024

Task Me Anything

Benchmarks for large multimodal language models (MLMs) now serve to simultaneously assess the general capabilities of models instead of evaluating for a specific capability. As a result, when a developer wants to identify which models to use for their application, they are overwhelmed by the number of benchmarks and remain uncertain about which benchmark's results are most reflective of their specific use case. This paper introduces Task-Me-Anything, a benchmark generation engine which produces a benchmark tailored to a user's needs. Task-Me-Anything maintains an extendable taxonomy of visual assets and can programmatically generate a vast number of task instances. Additionally, it algorithmically addresses user queries regarding MLM performance efficiently within a computational budget. It contains 113K images, 10K videos, 2K 3D object assets, over 365 object categories, 655 attributes, and 335 relationships. It can generate 750M image/video question-answering pairs, which focus on evaluating MLM perceptual capabilities. Task-Me-Anything reveals critical insights: open-source MLMs excel in object and attribute recognition but lack spatial and temporal understanding; each model exhibits unique strengths and weaknesses; larger models generally perform better, though exceptions exist; and GPT4o demonstrates challenges in recognizing rotating/moving objects and distinguishing colors.

cs.CV↗

arXiv2024

InsigHTable: Insight-driven Hierarchical Table Visualization with Reinforcement Learning

Embedding visual representations within original hierarchical tables can mitigate additional cognitive load stemming from the division of users' attention. The created hierarchical table visualizations can help users understand and explore complex data with multi-level attributes. However, because of many options available for transforming hierarchical tables and selecting subsets for embedding, the design space of hierarchical table visualizations becomes vast, and the construction process turns out to be tedious, hindering users from constructing hierarchical table visualizations with many data insights efficiently. We propose InsigHTable, a mixed-initiative and insight-driven hierarchical table transformation and visualization system. We first define data insights within hierarchical tables, which consider the hierarchical structure in the table headers. Since hierarchical table visualization construction is a sequential decision-making process, InsigHTable integrates a deep reinforcement learning framework incorporating an auxiliary rewards mechanism. This mechanism addresses the challenge of sparse rewards in constructing hierarchical table visualizations. Within the deep reinforcement learning framework, the agent continuously optimizes its decision-making process to create hierarchical table visualizations to uncover more insights by collaborating with analysts. We demonstrate the usability and effectiveness of InsigHTable through two case studies and sets of experiments. The results validate the effectiveness of the deep reinforcement learning framework and show that InsigHTable can facilitate users to construct hierarchical table visualizations and understand underlying data insights.

cs.HC↗

arXiv2024

Calibration of the Cryogenic Measurement System of a Resonant Haloscope Cavity

Possible light bosonic dark matter interactions with the Standard Model photon have been searched by microwave resonant cavities. In this paper, we demonstrate the cryogenic readout system calibration of a 7.138 GHz copper cavity with a loaded quality factor $Q_l=10^4$, operated at 22 mK temperature based on a dilution refrigerator. Our readout system consists of High Electron Mobility Transistors as cryogenic amplifiers at 4 K, plus room-temperature amplifiers and a spectrum analyzer for signal power detection. We test the system with a superconducting two-level system as a single-photon source in the microwave frequency regime and report an overall 95.6 dB system gain and -71.4 dB attenuation in the cavity's input channel. The effective noise temperature of the measurement system is 7.5 K.

hep-ex↗

arXiv2024

Demonstration of MaskSearch: Efficiently Querying Image Masks for Machine Learning Workflows

We demonstrate MaskSearch, a system designed to accelerate queries over databases of image masks generated by machine learning models. MaskSearch formalizes and accelerates a new category of queries for retrieving images and their corresponding masks based on mask properties, which support various applications, from identifying spurious correlations learned by models to exploring discrepancies between model saliency and human attention. This demonstration makes the following contributions:(1) the introduction of MaskSearch's graphical user interface (GUI), which enables interactive exploration of image databases through mask properties, (2) hands-on opportunities for users to explore MaskSearch's capabilities and constraints within machine learning workflows, and (3) an opportunity for conference attendees to understand how MaskSearch accelerates queries over image masks.

cs.DB↗

arXiv2024

Dark photon constraints from a 7.139 GHz cavity haloscope experiment

The dark photon is a promising candidate for the dark matter which comprises most of the matter in our visible Universe. Via kinetic mixing with the Standard Model it can also be resonantly converted to photons in an electromagnetic cavity, offering novel experimental possibilities for the discovery and study of dark matter. We report the results of a pathfinder dark photon dark matter cavity search experiment performed at Hunan Normal University and the Institute of Physics, Chinese Academy of Sciences, representing the first stage of the APEX (Axion and dark Photon EXperiment) program. Finding no statistically significant excess, we place an upper limit on the kinetic mixing parameter $|\chi|<3.7\times 10^{-13}$ around $m_A\simeq 29.5$ $\mu$eV at 90% confidence level. This result exceeds other constraints on dark photon dark matter in this frequency range by roughly an order of magnitude.

hep-ex↗

arXiv2024

Dynamics and Resonance Fluorescence from a Superconducting Artificial Atom Doubly Driven by Quantized and Classical Fields

We report an experimental demonstration of resonance fluorescence in a two-level superconducting artificial atom under two driving fields coupled to a detuned cavity. One of the fields is classical and the other is varied from quantum (vacuum fluctuations) to classical one by controlling the photon number inside the cavity. The device consists of a transmon qubit strongly coupled to a one-dimensional transmission line and a coplanar waveguide resonator. We observe a sideband anti-crossing and asymmetry in the emission spectra of the system through a one-dimensional transmission line, which is fundamentally different from the weak coupling case. By changing the photon number inside the cavity, the emission spectrum of our doubly driven system approaches to the case when the atom is driven by two classical bichromatic fields. We also measure the dynamical evolution of the system through the transmission line and study the properties of the first-order correlation function, Rabi oscillations and energy relaxation in the system. The study of resonance fluorescence from an atom driven by two fields promotes understanding decoherence in superconducting quantum circuits and may find applications in superconducting quantum computing and quantum networks.

quant-ph↗

arXiv2023

MaskSearch: Querying Image Masks at Scale

Machine learning tasks over image databases often generate masks that annotate image content (e.g., saliency maps, segmentation maps, depth maps) and enable a variety of applications (e.g., determine if a model is learning spurious correlations or if an image was maliciously modified to mislead a model). While queries that retrieve examples based on mask properties are valuable to practitioners, existing systems do not support them efficiently. In this paper, we formalize the problem and propose MaskSearch, a system that focuses on accelerating queries over databases of image masks while guaranteeing the correctness of query results. MaskSearch leverages a novel indexing technique and an efficient filter-verification query execution framework. Experiments with our prototype show that MaskSearch, using indexes approximately 5% of the compressed data size, accelerates individual queries by up to two orders of magnitude and consistently outperforms existing methods on various multi-query workloads that simulate dataset exploration and analysis processes.

cs.DB↗

arXiv2023

VOCALExplore: Pay-as-You-Go Video Data Exploration and Model Building [Technical Report]

We introduce VOCALExplore, a system designed to support users in building domain-specific models over video datasets. VOCALExplore supports interactive labeling sessions and trains models using user-supplied labels. VOCALExplore maximizes model quality by automatically deciding how to select samples based on observed skew in the collected labels. It also selects the optimal video representations to use when training models by casting feature selection as a rising bandit problem. Finally, VOCALExplore implements optimizations to achieve low latency without sacrificing model performance. We demonstrate that VOCALExplore achieves close to the best possible model quality given candidate acquisition functions and feature extractors, and it does so with low visible latency (~1 second per iteration) and no expensive preprocessing.

cs.DB↗

arXiv2023

EQUI-VOCAL: Synthesizing Queries for Compositional Video Events from Limited User Interactions [Technical Report]

We introduce EQUI-VOCAL: a new system that automatically synthesizes queries over videos from limited user interactions. The user only provides a handful of positive and negative examples of what they are looking for. EQUI-VOCAL utilizes these initial examples and additional ones collected through active learning to efficiently synthesize complex user queries. Our approach enables users to find events without database expertise, with limited labeling effort, and without declarative specifications or sketches. Core to EQUI-VOCAL's design is the use of spatio-temporal scene graphs in its data model and query language and a novel query synthesis approach that works on large and noisy video data. Our system outperforms two baseline systems -- in terms of F1 score, synthesis time, and robustness to noise -- and can flexibly synthesize complex queries that the baselines do not support.

cs.DB↗

arXiv2022

Share the Tensor Tea: How Databases can Leverage the Machine Learning Ecosystem

We demonstrate Tensor Query Processor (TQP): a query processor that automatically compiles relational operators into tensor programs. By leveraging tensor runtimes such as PyTorch, TQP is able to: (1) integrate with ML tools (e.g., Pandas for data ingestion, Tensorboard for visualization); (2) target different hardware (e.g., CPU, GPU) and software (e.g., browser) backends; and (3) end-to-end accelerate queries containing both relational and ML operators. TQP is generic enough to support the TPC-H benchmark, and it provides performance that is comparable to, and often better than, that of specialized CPU and GPU query processors.

cs.DB↗

arXiv2022

Query Processing on Tensor Computation Runtimes

The huge demand for computation in artificial intelligence (AI) is driving unparalleled investments in hardware and software systems for AI. This leads to an explosion in the number of specialized hardware devices, which are now offered by major cloud vendors. By hiding the low-level complexity through a tensor-based interface, tensor computation runtimes (TCRs) such as PyTorch allow data scientists to efficiently exploit the exciting capabilities offered by the new hardware. In this paper, we explore how database management systems can ride the wave of innovation happening in the AI space. We design, build, and evaluate Tensor Query Processor (TQP): TQP transforms SQL queries into tensor programs and executes them on TCRs. TQP is able to run the full TPC-H benchmark by implementing novel algorithms for relational operators on the tensor routines. At the same time, TQP can support various hardware while only requiring a fraction of the usual development effort. Experiments show that TQP can improve query execution time by up to 10$\times$ over specialized CPU- and GPU-only systems. Finally, TQP can accelerate queries mixing ML predictions and SQL end-to-end, and deliver up to 9$\times$ speedup over CPU baselines.

cs.DB↗

arXiv2021

DeepEverest: Accelerating Declarative Top-K Queries for Deep Neural Network Interpretation

We design, implement, and evaluate DeepEverest, a system for the efficient execution of interpretation by example queries over the activation values of a deep neural network. DeepEverest consists of an efficient indexing technique and a query execution algorithm with various optimizations. We prove that the proposed query execution algorithm is instance optimal. Experiments with our prototype show that DeepEverest, using less than 20% of the storage of full materialization, significantly accelerates individual queries by up to 63x and consistently outperforms other methods on multi-query workloads that simulate DNN interpretation processes.

cs.DB↗

arXiv2021

GSECnet: Ground Segmentation of Point Clouds for Edge Computing

Ground segmentation of point clouds remains challenging because of the sparse and unordered data structure. This paper proposes the GSECnet - Ground Segmentation network for Edge Computing, an efficient ground segmentation framework of point clouds specifically designed to be deployable on a low-power edge computing unit. First, raw point clouds are converted into a discretization representation by pillarization. Afterward, features of points within pillars are fed into PointNet to get the corresponding pillars feature map. Then, a depthwise-separable U-Net with the attention module learns the classification from the pillars feature map with an enormously diminished model parameter size. Our proposed framework is evaluated on SemanticKITTI against both point-based and discretization-based state-of-the-art learning approaches, and achieves an excellent balance between high accuracy and low computing complexity. Remarkably, our framework achieves the inference runtime of 135.2 Hz on a desktop platform. Moreover, experiments verify that it is deployable on a low-power edge computing unit powered 10 watts only.

cs.CV↗