arXiv subjects

Zhi Chen

Publications and source records attributed to Zhi Chen.

At least 145 records · Page 8Linked to original sources

arXiv2024

GUI Testing Arena: A Unified Benchmark for Advancing Autonomous GUI Testing Agent

Nowadays, research on GUI agents is a hot topic in the AI community. However, current research focuses on GUI task automation, limiting the scope of applications in various GUI scenarios. In this paper, we propose a formalized and comprehensive environment to evaluate the entire process of automated GUI Testing (GTArena), offering a fair, standardized environment for consistent operation of diverse multimodal large language models. We divide the testing process into three key subtasks: test intention generation, test task execution, and GUI defect detection, and construct a benchmark dataset based on these to conduct a comprehensive evaluation. It evaluates the performance of different models using three data types: real mobile applications, mobile applications with artificially injected defects, and synthetic data, thoroughly assessing their capabilities in this relevant task. Additionally, we propose a method that helps researchers explore the correlation between the performance of multimodal language large models in specific scenarios and their general capabilities in standard benchmark tests. Experimental results indicate that even the most advanced models struggle to perform well across all sub-tasks of automated GUI Testing, highlighting a significant gap between the current capabilities of Autonomous GUI Testing and its practical, real-world applicability. This gap provides guidance for the future direction of GUI Agent development. Our code is available at https://github.com/ZJU-ACES-ISE/ChatUITest.

cs.AI↗

arXiv2024

UAV-Enabled Passive 6D Movable Antennas: Joint Deployment and Beamforming Optimization

Intelligent reflecting surface (IRS) is composed of numerous passive reflecting elements and can be mounted on unmanned aerial vehicles (UAVs) to achieve six-dimensional (6D) movement by adjusting the UAV's three-dimensional (3D) location and 3D orientation simultaneously. Hence, in this paper, we investigate a new UAV-enabled passive 6D movable antenna (6DMA) architecture by mounting an IRS on a UAV and address the associated joint deployment and beamforming optimization problem. In particular, we consider a passive 6DMA-aided multicast system with a multi-antenna base station (BS) and multiple remote users, aiming to jointly optimize the IRS's location and 3D orientation, as well as its passive beamforming to maximize the minimum received signal-to-noise ratio (SNR) among all users under the practical angle-dependent signal reflection model. However, this optimization problem is challenging to be optimally solved due to the intricate relationship between the users' SNRs and the IRS's location and orientation. To tackle this challenge, we first focus on a simplified case with a single user, showing that one-dimensional (1D) orientation suffices to achieve the optimal performance. Next, we show that for any given IRS's location, the optimal 1D orientation can be derived in closed form, based on which several useful insights are drawn. To solve the max-min SNR problem in the general multi-user case, we propose an alternating optimization (AO) algorithm by alternately optimizing the IRS's beamforming and location/orientation via successive convex approximation (SCA) and hybrid coarse- and fine-grained search, respectively. To avoid undesirable local sub-optimal solutions, a Gibbs sampling (GS) method is proposed to generate new IRS locations and orientations for exploration in each AO iteration. Numerical results validate our theoretical analyses.

eess.SP↗

arXiv2024

On the Element-Wise Representation and Reasoning in Zero-Shot Image Recognition: A Systematic Survey

Zero-shot image recognition (ZSIR) aims to recognize and reason in unseen domains by learning generalized knowledge from limited data in the seen domain. The gist of ZSIR is constructing a well-aligned mapping between the input visual space and the target semantic space, which is a bottom-up paradigm inspired by the process by which humans observe the world. In recent years, ZSIR has witnessed significant progress on a broad spectrum, from theory to algorithm design, as well as widespread applications. However, to the best of our knowledge, there remains a lack of a systematic review of ZSIR from an element-wise perspective, i.e., learning fine-grained elements of data and their inferential associations. To fill the gap, this paper thoroughly investigates recent advances in element-wise ZSIR and provides a sound basis for its future development. Concretely, we first integrate three basic ZSIR tasks, i.e., object recognition, compositional recognition, and foundation model-based open-world recognition, into a unified element-wise paradigm and provide a detailed taxonomy and analysis of the main approaches. Next, we summarize the benchmarks, covering technical implementations, standardized datasets, and some more details as a library. Last, we sketch out related applications, discuss vital challenges, and suggest potential future directions.

cs.CV↗

arXiv2024

The Challenges of Modeling Astrophysical Reacting Flows

Stellar evolution is driven by the changing composition of a star from nuclear reactions. At the late stages of evolution and during explosive events, the timescale can be short and drive strong hydrodynamic flows, making simulations of astrophysical reacting flows challenging. Over the past decades, the standard approach to modeling reactions in simulation codes has been operator splitting, using implicit integrators for reactions. Here we explore some of the assumptions in this standard approach and describe some techniques for improving the efficiency and accuracy of astrophysical reacting flows.

astro-ph.IM↗

arXiv2024

Track Any Peppers: Weakly Supervised Sweet Pepper Tracking Using VLMs

In the Detection and Multi-Object Tracking of Sweet Peppers Challenge, we present Track Any Peppers (TAP) - a weakly supervised ensemble technique for sweet peppers tracking. TAP leverages the zero-shot detection capabilities of vision-language foundation models like Grounding DINO to automatically generate pseudo-labels for sweet peppers in video sequences with minimal human intervention. These pseudo-labels, refined when necessary, are used to train a YOLOv8 segmentation network. To enhance detection accuracy under challenging conditions, we incorporate pre-processing techniques such as relighting adjustments and apply depth-based filtering during post-inference. For object tracking, we integrate the Matching by Segment Anything (MASA) adapter with the BoT-SORT algorithm. Our approach achieves a HOTA score of 80.4%, MOTA of 66.1%, Recall of 74.0%, and Precision of 90.7%, demonstrating effective tracking of sweet peppers without extensive manual effort. This work highlights the potential of foundation models for efficient and accurate object detection and tracking in agricultural settings.

cs.CV↗

arXiv2024

FinCon: A Synthesized LLM Multi-Agent System with Conceptual Verbal Reinforcement for Enhanced Financial Decision Making

Large language models (LLMs) have demonstrated notable potential in conducting complex tasks and are increasingly utilized in various financial applications. However, high-quality sequential financial investment decision-making remains challenging. These tasks require multiple interactions with a volatile environment for every decision, demanding sufficient intelligence to maximize returns and manage risks. Although LLMs have been used to develop agent systems that surpass human teams and yield impressive investment returns, opportunities to enhance multi-sourced information synthesis and optimize decision-making outcomes through timely experience refinement remain unexplored. Here, we introduce the FinCon, an LLM-based multi-agent framework with CONceptual verbal reinforcement tailored for diverse FINancial tasks. Inspired by effective real-world investment firm organizational structures, FinCon utilizes a manager-analyst communication hierarchy. This structure allows for synchronized cross-functional agent collaboration towards unified goals through natural language interactions and equips each agent with greater memory capacity than humans. Additionally, a risk-control component in FinCon enhances decision quality by episodically initiating a self-critiquing mechanism to update systematic investment beliefs. The conceptualized beliefs serve as verbal reinforcement for the future agent's behavior and can be selectively propagated to the appropriate node that requires knowledge updates. This feature significantly improves performance while reducing unnecessary peer-to-peer communication costs. Moreover, FinCon demonstrates strong generalization capabilities in various financial tasks, including single stock trading and portfolio management.

cs.CL↗

arXiv2024

What Factors Affect Multi-Modal In-Context Learning? An In-Depth Exploration

Recently, rapid advancements in Multi-Modal In-Context Learning (MM-ICL) have achieved notable success, which is capable of achieving superior performance across various tasks without requiring additional parameter tuning. However, the underlying rules for the effectiveness of MM-ICL remain under-explored. To fill this gap, this work aims to investigate the research question: "What factors affect the performance of MM-ICL?'' To this end, we investigate extensive experiments on the three core steps of MM-ICL including demonstration retrieval, demonstration ordering, and prompt construction using 6 vision large language models and 20 strategies. Our findings highlight (1) the necessity of a multi-modal retriever for demonstration retrieval, (2) the importance of intra-demonstration ordering over inter-demonstration ordering, and (3) the enhancement of task comprehension through introductory instructions in prompts. We hope this study can serve as a foundational guide for optimizing MM-ICL strategies in future research.

cs.CL↗

arXiv2024

KVSharer: Efficient Inference via Layer-Wise Dissimilar KV Cache Sharing

The development of large language models (LLMs) has significantly expanded model sizes, resulting in substantial GPU memory requirements during inference. The key and value storage of the attention map in the KV (key-value) cache accounts for more than 80\% of this memory consumption. Nowadays, most existing KV cache compression methods focus on intra-layer compression within a single Transformer layer but few works consider layer-wise compression. In this paper, we propose a plug-and-play method called \textit{KVSharer}, which shares the KV cache between layers to achieve layer-wise compression. Rather than intuitively sharing based on higher similarity, we discover a counterintuitive phenomenon: sharing dissimilar KV caches better preserves the model performance. Experiments show that \textit{KVSharer} can reduce KV cache computation by 30\%, thereby lowering memory consumption without significantly impacting model performance and it can also achieve at least 1.3 times generation acceleration. Additionally, we verify that \textit{KVSharer} is compatible with existing intra-layer KV cache compression methods, and combining both can further save memory.

cs.LG↗

arXiv2024

Strong Coupling of Hydrodynamics and Reactions in Nuclear Statistical Equilibrium for Modeling Convection in Massive Stars

We build on the simplified spectral deferred corrections (SDC) coupling of hydrodynamics and reactions to handle the case of nuclear statistical equilibrium (NSE) and electron/positron captures/decays in the cores of massive stars. Our approach blends a traditional reaction network on the grid with a tabulated NSE state from a very large, O(100) nuclei, network. We demonstrate how to achieve second-order accuracy in the simplified-SDC framework when coupling NSE to hydrodynamics, with the ability to evolve the star on the hydrodynamics timestep. We discuss the application of this method to convection in massive stars leading up to core-collapse. We also show how to initialize the initial convective state from a 1D model in a self-consistent fashion. All of these developments are done in the publicly available Castro simulation code and the entire simulation methodology is fully GPU accelerated.

astro-ph.SR↗

arXiv2024

Codebook Design and Performance Analysis for Wideband Beamforming in Terahertz Communications

The codebook-based analog beamforming is appealing for future terahertz (THz) communications since it can generate high-gain directional beams with low-cost phase shifters via low-complexity beam training. However, conventional beamforming codebook design based on array response vectors for narrowband communications may suffer from severe performance loss in wideband systems due to the ``beam squint" effect over frequency. To tackle this issue, we propose in this paper a new codebook design method for analog beamforming in wideband THz systems. In particular, to characterize the analog beamforming performance in wideband systems, we propose a new metric termed wideband beam gain, which is given by the minimum beamforming gain over the entire frequency band given a target angle. Based on this metric, a wideband analog beamforming codebook design problem is formulated for optimally balancing the beamforming gains in both the spatial and frequency domains, and the performance loss of conventional narrowband beamforming in wideband systems is analyzed. To solve the new wideband beamforming codebook design problem, we divide the spatial domain into orthogonal angular zones each associated with one beam, thereby decoupling the codebook design into a zone division sub-problem and a set of beamforming optimization sub-problems each for one zone. For the zone division sub-problem, we propose a bisection method to obtain the optimal boundaries for separating adjacent zones. While for each of the per-zone-based beamforming optimization sub-problems, we further propose an efficient augmented Lagrange method (ALM) to solve it. Numerical results demonstrate the performance superiority of our proposed codebook design for wideband analog beamforming to the narrowband beamforming codebook and also validate our performance analysis.

eess.SP↗

arXiv2024

TurboRAG: Accelerating Retrieval-Augmented Generation with Precomputed KV Caches for Chunked Text

Current Retrieval-Augmented Generation (RAG) systems concatenate and process numerous retrieved document chunks for prefill which requires a large volume of computation, therefore leading to significant latency in time-to-first-token (TTFT). To reduce the computation overhead as well as TTFT, we introduce TurboRAG, a novel RAG system that redesigns the inference paradigm of the current RAG system by first pre-computing and storing the key-value (KV) caches of documents offline, and then directly retrieving the saved KV cache for prefill. Hence, online computation of KV caches is eliminated during inference. In addition, we provide a number of insights into the mask matrix and positional embedding mechanisms, plus fine-tune a pretrained language model to maintain model accuracy of TurboRAG. Our approach is applicable to most existing large language models and their applications without any requirement in modification of models and inference systems. Experimental results across a suite of RAG benchmarks demonstrate that TurboRAG reduces TTFT by up to 9.4x compared to the conventional RAG systems (on an average of 8.6x), but reserving comparable performance to the standard RAG systems.

cs.CV↗

arXiv2024

TimeCNN: Refining Cross-Variable Interaction on Time Point for Time Series Forecasting

Time series forecasting is extensively applied across diverse domains. Transformer-based models demonstrate significant potential in modeling cross-time and cross-variable interaction. However, we notice that the cross-variable correlation of multivariate time series demonstrates multifaceted (positive and negative correlations) and dynamic progression over time, which is not well captured by existing Transformer-based models. To address this issue, we propose a TimeCNN model to refine cross-variable interactions to enhance time series forecasting. Its key innovation is timepoint-independent, where each time point has an independent convolution kernel, allowing each time point to have its independent model to capture relationships among variables. This approach effectively handles both positive and negative correlations and adapts to the evolving nature of variable relationships over time. Extensive experiments conducted on 12 real-world datasets demonstrate that TimeCNN consistently outperforms state-of-the-art models. Notably, our model achieves significant reductions in computational requirements (approximately 60.46%) and parameter count (about 57.50%), while delivering inference speeds 3 to 4 times faster than the benchmark iTransformer model

cs.LG↗

arXiv2024

Patient-Specific CT Doses Using DL-based Image Segmentation and GPU-based Monte Carlo Calculations for 10,281 Subjects

Computed tomography (CT) scans are a major source of medical radiation exposure worldwide. In countries like China, the frequency of CT scans has grown rapidly, particularly in routine physical examinations where chest CT scans are increasingly common. Accurate estimation of organ doses is crucial for assessing radiation risk and optimizing imaging protocols. However, traditional methods face challenges due to the labor-intensive process of manual organ segmentation and the computational demands of Monte Carlo (MC) dose calculations. In this study, we present a novel method that combines automatic image segmentation with GPU-accelerated MC simulations to compute patient-specific organ doses for a large cohort of 10,281 individuals undergoing CT examinations for physical examinations at a Chinese hospital. This is the first big-data study of its kind involving such a large population for CT dosimetry. The results show considerable inter-individual variability in CTDIvol-normalized organ doses, even among subjects with similar BMI or WED. Patient-specific organ doses vary widely, ranging from 33% to 164% normalized by the doses from ICRP Adult Reference Phantoms. Statistical analyses indicate that the "Reference Man" based average phantoms can lead to significant dosimetric uncertainties, with relative errors exceeding 50% in some cases. These findings underscore the fact that previous assessments of radiation risk may be inaccurate. It took our computational tool, on average, 135 seconds per subject, using a single NVIDIA RTX 3080 GPU card. The big-data analysis provides interesting data for improving CT dosimetry and risk assessment by avoiding uncertainties that were neglected in the past.

physics.med-ph↗

arXiv2024

Promise and Peril of Collaborative Code Generation Models: Balancing Effectiveness and Memorization

In the rapidly evolving field of machine learning, training models with datasets from various locations and organizations presents significant challenges due to privacy and legal concerns. The exploration of effective collaborative training settings capable of leveraging valuable knowledge from distributed and isolated datasets is increasingly crucial. This study investigates key factors that impact the effectiveness of collaborative training methods in code next-token prediction, as well as the correctness and utility of the generated code, demonstrating the promise of such methods. Additionally, we evaluate the memorization of different participant training data across various collaborative training settings, including centralized, federated, and incremental training, highlighting their potential risks in leaking data. Our findings indicate that the size and diversity of code datasets are pivotal factors influencing the success of collaboratively trained code models. We show that federated learning achieves competitive performance compared to centralized training while offering better data protection, as evidenced by lower memorization ratios in the generated code. However, federated learning can still produce verbatim code snippets from hidden training data, potentially violating privacy or copyright. Our study further explores effectiveness and memorization patterns in incremental learning, emphasizing the sequence in which individual participant datasets are introduced. We also identify cross-organizational clones as a prevalent challenge in both centralized and federated learning scenarios. Our findings highlight the persistent risk of data leakage during inference, even when training data remains unseen. We conclude with recommendations for practitioners and researchers to optimize multisource datasets, propelling cross-organizational collaboration forward.

cs.SE↗

arXiv2024

CF-PRNet: Coarse-to-Fine Prototype Refining Network for Point Cloud Completion and Reconstruction

In modern agriculture, precise monitoring of plants and fruits is crucial for tasks such as high-throughput phenotyping and automated harvesting. This paper addresses the challenge of reconstructing accurate 3D shapes of fruits from partial views, which is common in agricultural settings. We introduce CF-PRNet, a coarse-to-fine prototype refining network, leverages high-resolution 3D data during the training phase but requires only a single RGB-D image for real-time inference. Our approach begins by extracting the incomplete point cloud data that constructed from a partial view of a fruit with a series of convolutional blocks. The extracted features inform the generation of scaling vectors that refine two sequentially constructed 3D mesh prototypes - one coarse and one fine-grained. This progressive refinement facilitates the detailed completion of the final point clouds, achieving detailed and accurate reconstructions. CF-PRNet demonstrates excellent performance metrics with a Chamfer Distance of 3.78, an F1 Score of 66.76%, a Precision of 56.56%, and a Recall of 85.31%, and win the first place in the Shape Completion and Reconstruction of Sweet Peppers Challenge.

cs.CV↗

arXiv2024

FiAt-Net: Detecting Fibroatheroma Plaque Cap in 3D Intravascular OCT Images

The key manifestation of coronary artery disease (CAD) is development of fibroatheromatous plaque, the cap of which may rupture and subsequently lead to coronary artery blocking and heart attack. As such, quantitative analysis of coronary plaque, its plaque cap, and consequently the cap's likelihood to rupture are of critical importance when assessing a risk of cardiovascular events. This paper reports a new deep learning based approach, called FiAt-Net, for detecting angular extent of fibroatheroma (FA) and segmenting its cap in 3D intravascular optical coherence tomography (IVOCT) images. IVOCT 2D image frames are first associated with distinct clusters and data from each cluster are used for model training. As plaque is typically focal and thus unevenly distributed, a binary partitioning method is employed to identify FA plaque areas to focus on to mitigate the data imbalance issue. Additional image representations (called auxiliary images) are generated to capture IVOCT intensity changes to help distinguish FA and non-FA areas on the coronary wall. Information in varying scales is derived from the original IVOCT and auxiliary images, and a multi-head self-attention mechanism is employed to fuse such information. Our FiAt-Net achieved high performance on a 3D IVOCT coronary image dataset, demonstrating its effectiveness in accurately detecting FA cap in IVOCT images.

eess.IV↗

arXiv2024

PlantSeg: A Large-Scale In-the-wild Dataset for Plant Disease Segmentation

Plant diseases pose significant threats to agriculture. It necessitates proper diagnosis and effective treatment to safeguard crop yields. To automate the diagnosis process, image segmentation is usually adopted for precisely identifying diseased regions, thereby advancing precision agriculture. Developing robust image segmentation models for plant diseases demands high-quality annotations across numerous images. However, existing plant disease datasets typically lack segmentation labels and are often confined to controlled laboratory settings, which do not adequately reflect the complexity of natural environments. Motivated by this fact, we established PlantSeg, a large-scale segmentation dataset for plant diseases. PlantSeg distinguishes itself from existing datasets in three key aspects. (1) Annotation type: Unlike the majority of existing datasets that only contain class labels or bounding boxes, each image in PlantSeg includes detailed and high-quality segmentation masks, associated with plant types and disease names. (2) Image source: Unlike typical datasets that contain images from laboratory settings, PlantSeg primarily comprises in-the-wild plant disease images. This choice enhances the practical applicability, as the trained models can be applied for integrated disease management. (3) Scale: PlantSeg is extensive, featuring 11,400 images with disease segmentation masks and an additional 8,000 healthy plant images categorized by plant type. Extensive technical experiments validate the high quality of PlantSeg's annotations. This dataset not only allows researchers to evaluate their image classification methods but also provides a critical foundation for developing and benchmarking advanced plant disease segmentation algorithms.

cs.CV↗

arXiv2024

Movable Antennas Meet Intelligent Reflecting Surface: When Do We Need Movable Antennas?

Intelligent reflecting surface (IRS) and movable antenna (MA)/fluid antenna (FA) techniques have both received increasing attention in the realm of wireless communications due to their ability to reconfigure and improve wireless channel conditions. In this paper, we investigate the integration of MAs/FAs into an IRS-assisted wireless communication system. In particular, we consider the downlink transmission from a multi-MA base station (BS) to a single-antenna user with the aid of an IRS, aiming to maximize the user's received signal-to-noise ratio (SNR), by jointly optimizing the BS/IRS active/passive beamforming and the MAs' positions. Due to the similar capability of MAs and IRS for channel reconfiguration, we first conduct theoretical analyses of the performance gain of MAs over conventional fixed-position antennas (FPAs) under the line-of-sight (LoS) BS-IRS channel and derive the conditions under which the performance gain becomes more or less significant. Next, to solve the received SNR maximization problem, we propose an alternating optimization (AO) algorithm that decomposes it into two subproblems and solve them alternately. Numerical results are provided to validate our analytical results and evaluate the performance gains of MAs over FPAs under different setups.

cs.IT↗