arXiv subjects

Yujie Wang

Publications and source records attributed to Yujie Wang.

At least 91 records · Page 5Linked to original sources

arXiv2024

Dynamic constraints predict the relaxation of granular materials

Granular materials such as sand, powders, and food grains are ubiquitous in civil engineering, geoscience, agriculture, and medicine. While the influence of friction between the grains on the static structure of these systems is well understood, its impact on the dynamics is an open problem. Here we use particle-based simulations of a granular pack under cyclic shear and discover that the relaxation time of the system is a non-monotonic function of friction. By introducing the concept of dynamic constraints, we reveal that this re-entrant dynamics is due to the competition between increasing frictional coupling and a concurrent change in the structure of the granular pack. Our theoretical approach, which unifies the dynamics of friction-less systems with frictional ones, is applicable to other systems that have a complex free energy landscape and a dynamics which involves time-dependent constraints, thus setting the stage for a description of the dynamic behavior of a large class of complex systems.

cond-mat.soft↗

arXiv2024

Dynamic Activation Pitfalls in LLaMA Models: An Empirical Study

In this work, we systematically investigate the efficacy of dynamic activation mechanisms within the LLaMA family of language models. Despite the potential of dynamic activation methods to reduce computation and increase speed in models using the ReLU activation function, our empirical findings have uncovered several inherent pitfalls in the current dynamic activation schemes. Through extensive experiments across various dynamic activation strategies, we demonstrate that LLaMA models usually underperform when compared to their ReLU counterparts, particularly in scenarios demanding high sparsity ratio. We attribute these deficiencies to a combination of factors: 1) the inherent complexity of dynamically predicting activation heads and neurons; 2) the inadequate sparsity resulting from activation functions; 3) the insufficient preservation of information resulting from KV cache skipping. Our analysis not only sheds light on the limitations of dynamic activation in the context of large-scale LLaMA models but also proposes roadmaps for enhancing the design of future sparsity schemes.

cs.LG↗

arXiv2024

Adaptive Safety-Critical Control for a Class of Nonlinear Systems with Parametric Uncertainties: A Control Barrier Function Approach

This paper presents a novel approach for the safe control design of systems with parametric uncertainties in both drift terms and control-input matrices. The method combines control barrier functions and adaptive laws to generate a safe controller through a nonlinear program with an explicitly given closed-form solution. The proposed approach verifies the non-emptiness of the admissible control set independently of online parameter estimations, which can ensure the safe controller is singularity-free. A data-driven algorithm is also developed to improve the performance of the proposed controller by tightening the bounds of the unknown parameters. The effectiveness of the control scheme is demonstrated through numerical simulations.

eess.SY↗

arXiv2024

Near-field reconstruction of periodic structures with superimposed illumination

This paper addresses the reconstruction of periodic structures using phase or phaseless near-field measurements. We introduce a novel illumination strategy based on the quasi-periodic condition. Employing the Dirichlet-to-Neumann (DtN) map, we reformulate the scattering problem into a uniquely solvable boundary value problem. Our theoretical analysis shows the advantages of employing a fixed wavenumber superposition illumination strategy and proves that a specific wavenumber interval and fixed direction uniquely determine the grating structure. Numerical results confirm the efficacy of this illumination strategy and reconstruction algorithm, enabling super-resolution imaging from both phase and phaseless near-field measurements.

math-ph↗

arXiv2024

Cross-domain Chinese Sentence Pattern Parsing

Sentence Pattern Structure (SPS) parsing is a syntactic analysis method primarily employed in language teaching.Existing SPS parsers rely heavily on textbook corpora for training, lacking cross-domain capability.To overcome this constraint, this paper proposes an innovative approach leveraging large language models (LLMs) within a self-training framework. Partial syntactic rules from a source domain are combined with target domain sentences to dynamically generate training data, enhancing the adaptability of the parser to diverse domains.Experiments conducted on textbook and news domains demonstrate the effectiveness of the proposed method, outperforming rule-based baselines by 1.68 points on F1 metrics.

cs.CL↗

arXiv2024

Facile synthesis of micro-flower NiCo2O4 assembled by nanosheets efficient for electrocatalysis of water

Effective regulation of the morphology of transition metal spinel structures is crucial for creating efficient and stable bifunctional catalysts for electrocatalysis of water. In this work, micro-flower NiCo2O4 (F-NCO) assembled by nanosheets via a chemical template method for the simultaneous promotion of hydrogen evolution reaction (HER) and oxygen evolution reaction (OER). Electronic microscope analysis revealed that the thickness of the F-NCO catalyst was only 2.7% of that of the NiCo2O4 bulk (B-NCO), and this ultrathin lamellar structure was conducive to further exposure of the active site and improved reaction kinetics. The F-NCO catalyst exhibited superior HER and OER performance (10 = 236 and 310 mV) and robust long-term stability over the B-NCO catalyst in 1.0 M KOH, with a 2.68-fold and 4.16-fold increase in active surface area and a 0.42-fold and 0.61-fold decrease in charge transfer resistance values, respectively. This micro-flower-structured electrode has remarkable electrocatalytic property and long-term durability, providing a novel insight for characterizing cost-effective and high-performance bifunctional electrocatalysts.

physics.chem-ph↗

arXiv2024

Improving Automatic Parallel Training via Balanced Memory Workload Optimization

Transformer models have emerged as the leading approach for achieving state-of-the-art performance across various application domains, serving as the foundation for advanced large-scale deep learning (DL) models. However, efficiently training these models across multiple GPUs remains a complex challenge due to the abundance of parallelism options. Existing DL systems either require manual efforts to design distributed training plans or limit parallelism combinations to a constrained search space. In this paper, we present Galvatron-BMW, a novel system framework that integrates multiple prevalent parallelism dimensions and automatically identifies the most efficient hybrid parallelism strategy. To effectively navigate this vast search space, we employ a decision tree approach for decomposition and pruning based on intuitive insights. We further utilize a dynamic programming search algorithm to derive the optimal plan. Moreover, to improve resource utilization and enhance system efficiency, we propose a bi-objective optimization workflow that focuses on workload balance. Our evaluations on different Transformer models demonstrate the capabilities of Galvatron-BMW in automating distributed training under varying GPU memory constraints. Across all tested scenarios, Galvatron-BMW consistently achieves superior system throughput, surpassing previous approaches that rely on limited parallelism strategies.

cs.LG↗

arXiv2024

From Text to CQL: Bridging Natural Language and Corpus Search Engine

Natural Language Processing (NLP) technologies have revolutionized the way we interact with information systems, with a significant focus on converting natural language queries into formal query languages such as SQL. However, less emphasis has been placed on the Corpus Query Language (CQL), a critical tool for linguistic research and detailed analysis within text corpora. The manual construction of CQL queries is a complex and time-intensive task that requires a great deal of expertise, which presents a notable challenge for both researchers and practitioners. This paper presents the first text-to-CQL task that aims to automate the translation of natural language into CQL. We present a comprehensive framework for this task, including a specifically curated large-scale dataset and methodologies leveraging large language models (LLMs) for effective text-to-CQL task. In addition, we established advanced evaluation metrics to assess the syntactic and semantic accuracy of the generated queries. We created innovative LLM-based conversion approaches and detailed experiments. The results demonstrate the efficacy of our methods and provide insights into the complexities of text-to-CQL task.

cs.CL↗

arXiv2024

Weaver: Foundation Models for Creative Writing

This work introduces Weaver, our first family of large language models (LLMs) dedicated to content creation. Weaver is pre-trained on a carefully selected corpus that focuses on improving the writing capabilities of large language models. We then fine-tune Weaver for creative and professional writing purposes and align it to the preference of professional writers using a suit of novel methods for instruction data synthesis and LLM alignment, making it able to produce more human-like texts and follow more diverse instructions for content creation. The Weaver family consists of models of Weaver Mini (1.8B), Weaver Base (6B), Weaver Pro (14B), and Weaver Ultra (34B) sizes, suitable for different applications and can be dynamically dispatched by a routing agent according to query complexity to balance response quality and computation cost. Evaluation on a carefully curated benchmark for assessing the writing capabilities of LLMs shows Weaver models of all sizes outperform generalist LLMs several times larger than them. Notably, our most-capable Weaver Ultra model surpasses GPT-4, a state-of-the-art generalist LLM, on various writing scenarios, demonstrating the advantage of training specialized LLMs for writing purposes. Moreover, Weaver natively supports retrieval-augmented generation (RAG) and function calling (tool usage). We present various use cases of these abilities for improving AI-assisted writing systems, including integration of external knowledge bases, tools, or APIs, and providing personalized writing assistance. Furthermore, we discuss and summarize a guideline and best practices for pre-training and fine-tuning domain-specific LLMs.

cs.CL↗

arXiv2023

Quantum-Enhanced Forecasting: Leveraging Quantum Gramian Angular Field and CNNs for Stock Return Predictions

We propose a time series forecasting method named Quantum Gramian Angular Field (QGAF). This approach merges the advantages of quantum computing technology with deep learning, aiming to enhance the precision of time series classification and forecasting. We successfully transformed stock return time series data into two-dimensional images suitable for Convolutional Neural Network (CNN) training by designing specific quantum circuits. Distinct from the classical Gramian Angular Field (GAF) approach, QGAF's uniqueness lies in eliminating the need for data normalization and inverse cosine calculations, simplifying the transformation process from time series data to two-dimensional images. To validate the effectiveness of this method, we conducted experiments on datasets from three major stock markets: the China A-share market, the Hong Kong stock market, and the US stock market. Experimental results revealed that compared to the classical GAF method, the QGAF approach significantly improved time series prediction accuracy, reducing prediction errors by an average of 25% for Mean Absolute Error (MAE) and 48% for Mean Squared Error (MSE). This research confirms the potential and promising prospects of integrating quantum computing with deep learning techniques in financial time series forecasting.

cs.LG↗

arXiv2023

Experimental Study of the Nematic Transition in Granular Spherocylinder Packings under Tapping

Using x-ray tomography, we experimentally investigate the nematic transition in granular spherocylinder packings induced by tapping. Upon the validation of the Edwards ensemble framework in spherocylinders, we introduce an empirical free energy that accounts for the influence of gravity and the mechanical stability requirements specific to granular systems. This free energy can predict not only the correct phase transition behavior of the system from a disordered state to a nematic phase, but also a phase coexistence range and nucleation energy barriers that agree with experimental observations.

cond-mat.soft↗

arXiv2023

Experimental Study of Granular Clogging in Two-Dimensional Hopper

We experimentally investigate the clogging process of granular materials in a two-dimensional hopper, and present a self-consistent physical mechanism of clogging based on preformed dynamic chain structures in the flow. We found that these chain structures follow a specific modified restricted random walk, and clogging occurs when they are mechanically stable enough to withstand the flow fluctuations, resulting in the formation of an arch at the outlet. We introduce a simple model which can explain the clogging probability by incorporating an analytical expression for chain formation and its transition into an arch. Our results provide insight into the microscopic mechanism of clogging in hopper flow.

cond-mat.soft↗

arXiv2023

Safe Control of Euler-Lagrange Systems with Limited Model Information

This paper presents a new safe control framework for Euler-Lagrange (EL) systems with limited model information, external disturbances, and measurement uncertainties. The EL system is decomposed into two subsystems called the proxy subsystem and the virtual tracking subsystem. An adaptive safe controller based on barrier Lyapunov functions is designed for the virtual tracking subsystem to ensure the boundedness of the safe velocity tracking error, and a safe controller based on control barrier functions is designed for the proxy subsystem to ensure controlled invariance of the safe set defined either in the joint space or task space. Theorems that guarantee the safety of the proposed controllers are provided. In contrast to existing safe control strategies for EL systems, the proposed method requires much less model information and can ensure safety rather than input-to-state safety. Simulation results are provided to illustrate the effectiveness of the proposed method.

eess.SY↗

arXiv2023

Single image reflection removal via learning with multi-image constraints

Reflections are very common phenomena in our daily photography, which distract people's attention from the scene behind the glass. The problem of removing reflection artifacts is important but challenging due to its ill-posed nature. The traditional approaches solve an optimization problem over the constraints induced from multiple images, at the expense of large computation costs. Recent learning-based approaches have demonstrated a significant improvement in both performance and running time for single image reflection removal, but are limited as they require a large number of synthetic reflection/clean image pairs for direct supervision to approximate the ground truth, at the risk of overfitting in the synthetic image domain and degrading in the real image domain. In this paper, we propose a novel learning-based solution that combines the advantages of the aforementioned approaches and overcomes their drawbacks. Our algorithm works by learning a deep neural network to optimize the target with joint constraints enhanced among multiple input images during the training phase, but is able to eliminate reflections only from a single input for evaluation. Our algorithm runs in real-time and achieves state-of-the-art reflection removal performance on real images. We further propose a strong network backbone that disentangles the background and reflection information into separate latent codes, which are embedded into a shared one-branch deep neural network for both background and reflection predictions. The proposed backbone experimentally performs better than the other common network implementations, and provides insightful knowledge to understand the reflection removal task.

cs.CV↗

arXiv2023

Revisiting Prompt Engineering via Declarative Crowdsourcing

Large language models (LLMs) are incredibly powerful at comprehending and generating data in the form of text, but are brittle and error-prone. There has been an advent of toolkits and recipes centered around so-called prompt engineering-the process of asking an LLM to do something via a series of prompts. However, for LLM-powered data processing workflows, in particular, optimizing for quality, while keeping cost bounded, is a tedious, manual process. We put forth a vision for declarative prompt engineering. We view LLMs like crowd workers and leverage ideas from the declarative crowdsourcing literature-including leveraging multiple prompting strategies, ensuring internal consistency, and exploring hybrid-LLM-non-LLM approaches-to make prompt engineering a more principled process. Preliminary case studies on sorting, entity resolution, and imputation demonstrate the promise of our approach

cs.DB↗

arXiv2023

Safe Adaptive Multi-Agent Coverage Control

This paper presents a safe adaptive coverage controller for multi-agent systems with actuator faults and time-varying uncertainties. The centroidal Voronoi tessellation (CVT) is applied to generate an optimal configuration of multi-agent systems for covering an area of interest. As a conventional CVT-based controller cannot prevent collisions between agents with non-zero size, a control barrier function (CBF) based controller is developed to ensure collision avoidance with a function approximation technique (FAT) based design to deal with system uncertainties. The proposed controller is verified under simulations.

math.OC↗

arXiv2023

PolarDB-IMCI: A Cloud-Native HTAP Database System at Alibaba

Cloud-native databases have become the de-facto choice for mission-critical applications on the cloud due to the need for high availability, resource elasticity, and cost efficiency. Meanwhile, driven by the increasing connectivity between data generation and analysis, users prefer a single database to efficiently process both OLTP and OLAP workloads, which enhances data freshness and reduces the complexity of data synchronization and the overall business cost. In this paper, we summarize five crucial design goals for a cloud-native HTAP database based on our experience and customers' feedback, i.e., transparency, competitive OLAP performance, minimal perturbation on OLTP workloads, high data freshness, and excellent resource elasticity. As our solution to realize these goals, we present PolarDB-IMCI, a cloud-native HTAP database system designed and deployed at Alibaba Cloud. Our evaluation results show that PolarDB-IMCI is able to handle HTAP efficiently on both experimental and production workloads; notably, it speeds up analytical queries up to $\times149$ on TPC-H (100 $GB$). PolarDB-IMCI introduces low visibility delay and little performance perturbation on OLTP workloads (< 5%), and resource elasticity can be achieved by scaling out in tens of seconds.

cs.DB↗

arXiv2023

Disturbance Observer-based Robust Control Barrier Functions

This work presents a safe control design approach that integrates the disturbance observer (DOB) and the control barrier function (CBF) for systems with external disturbances. Different from existing robust CBF results that consider the "worst case" of disturbances, this work utilizes a DOB to estimate and compensate for the disturbances. DOB-CBF-based controllers are constructed with provably safe guarantees by solving convex quadratic programs online, to achieve a better tradeoff between safety and performance. Two types of systems are considered individually depending on the magnitude of the input and disturbance relative degrees. The effectiveness of the proposed methods is illustrated via numerical simulations.

eess.SY↗