检索结果-内蒙古大学图书馆

Beyond Pass or Fail: Multi-Dimensional Benchmarking of Foundation Models for Goal-based Mobile UI Navigation

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Ran, Dezhi Wu, Mengzhou Yu, Hao Li, Yuetong Ren, Jun Cao, Yuan Zeng, Xia Lu, Haochuan Xu, Zexin Xu, Mengqian Su, Ting Yao, Liangchao Xiong, Ting Yang, Wei Deng, Yuetang Marron, Assaf Harel, David Xie, Tao MOE SCS Peking University Beijing China School of EECS Peking University Beijing China School of Software and Microelectronics Peking University Beijing China The University of Chicago Chicago United States University of Texas at Dallas Dallas United States Tencent Inc Shenzhen China East China Normal University Shanghai China Dept. of Computer Science and Applied Mathematics Weizmann Institute of Science Rehovot Israel

Recent advances of foundation models (FMs) have made navigating mobile applications (apps) based on high-level goal instructions within reach, with significant industrial applications such as UI testing. While existing benchmarks evaluate FM-based UI navigation using the binary pass/fail metric, they have two major limitations: they cannot reflect the complex nature of mobile UI navigation where FMs may fail for various reasons (e.g., misunderstanding instructions and failed planning), and they lack industrial relevance due to oversimplified tasks that poorly represent real-world scenarios. To address the preceding limitations, we propose Sphinx, a comprehensive benchmark for multi-dimensional evaluation of FMs in industrial settings of UI navigation. Sphinx introduces a specialized toolkit that evaluates five essential FM capabilities, providing detailed insights into failure modes such as insufficient app knowledge or planning issues. Using both popular Google Play applications and WeChat’s internal UI test cases, we evaluate 8 FMs with 20 different configurations. Our results show that existing FMs universally struggle with goal-based testing tasks, primarily due to insufficient UI-specific capabilities. We summarize seven lessons learned from benchmarking FMs with Sphinx, providing clear directions for improving FM-based mobile UI navigation. Copyright © 2025, The Authors. All rights reserved.

关键词： Model checking

Real-World Image Super Resolution via Unsupervised Bi-directional Cycle Domain Transfer Learning based Generative Adversarial Network

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Wang, Xiang Yang, Yimin Guo, Zhichang Zhou, Zhili Liu, Yu Pang, Qixiang Du, Shan The Dept. of Computer Science Lakehead University Thunder BayON Canada The Dept. of electrical and computer engineering Western University LondonON Canada The School of Mathematics Harbin Institute of Technology Hei Long Jiang China The School of Computer and Software Nanjing University of Information Science and Technology Jiangsu China The School of Microelectronics Tianjin University Tianjin China The Okanagan College KelownaBC Canada The Dept. of Comp. Sci. Math Physics & Statistics The University of British Columbia OkanaganBC Canada

Deep Convolutional Neural Networks (DCNNs) have exhibited impressive performance on image super-resolution tasks. However, these deep learning-based super-resolution methods perform poorly in real-world super-resolution tasks, where the paired high-resolution and low-resolution images are unavailable and the low-resolution images are degraded by complicated and unknown kernels. To break these limitations, we propose the Unsupervised Bi-directional Cycle Domain Transfer Learning-based Generative Adversarial Network (UBCDTL-GAN), which consists of an Unsupervised Bi-directional Cycle Domain Transfer Network (UBCDTN) and the Semantic Encoder guided Super Resolution Network (SESRN). First, the UBCDTN is able to produce an approximated real-like LR image through transferring the LR image from an artificially degraded domain to the real-world LR image domain. Second, the SESRN has the ability to super-resolve the approximated real-like LR image to a photo-realistic HR image. Extensive experiments on unpaired real-world image benchmark datasets demonstrate that the proposed method achieves superior performance compared to state-of-the-art methods. © 2022, CC BY.

关键词： Convolutional neural networks

Finite impulse response filters for simplicial complexes

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Yang, Maosheng Isufi, Elvin Schaub, Michael T. Leus, Geert Dept. of Intelligent Systems Delft University of Technology Delft Netherlands Dept. of Computer Science RWTH Aachen University Aachen Germany Dept. of Microelectronics Delft University of Technology Delft Netherlands

In this paper, we study linear filters to process signals defined on simplicial complexes, i.e., signals defined on nodes, edges, triangles, etc. of a simplicial complex, thereby generalizing filtering operations for graph signals. We propose a finite impulse response filter based on the Hodge Laplacian, and demonstrate how this filter can be designed to amplify or attenuate certain spectral components of simplicial signals. Specifically, we discuss how, unlike in the case of node signals, the Fourier transform in the context of edge signals can be understood in terms of two orthogonal subspaces corresponding to the gradient-flow signals and curl-flow signals arising from the Hodge decomposition. By assigning different filter coefficients to the associated terms of the Hodge Laplacian, we develop a subspace-varying filter which enables more nuanced control over these signal types. Numerical experiments are conducted to show the potential of simplicial filters for sub-component extraction, denoising and model approximation. Copyright © 2021, The Authors. All rights reserved.

关键词： Laplace transforms

Language Modeling on a SpiNNaker2 Neuromorphic Chip

学校读者我要写书评

暂无评论

Language Modeling on a SpiNNaker2 Neuromorphic Chip

IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS)

作者： Khaleelulla Khan Nazeer Mark Schöne Rishav Mukherji Bernhard Vogginger Christian Mayr David Kappel Anand Subramoney Chair of Highly-Parallel VLSI-Systems and Neuro-Microelectronics Technische Universität Dresden Germany Birla Institute of Technology and Science Goa India Centre for Tactile Internet (CeTI) With Human-in-the-Loop Technische Universität Dresden Germany Institut für Neuroinformatik Ruhr Universität Bochum Bochum Germany Dept. of Computer Science Royal Holloway University of London Egham United Kingdom

ISBN: (数字)9798350383638

ISBN: (纸本)9798350383645

As large language models continue to scale in size rapidly, so too does the computational power required to run them. Event-based networks on neuromorphic devices offer a potential way to reduce energy consumption for inference significantly. However, to date, most event-based networks that can run on neuromorphic hardware, including spiking neural networks (SNNs), have not achieved task performance even on par with LSTM models for language modeling. As a result, language modeling on neuromorphic devices has seemed a distant prospect. In this work, we demonstrate the first-ever implementation of a language model on a neuromorphic device – specifically the SpiNNaker2 chip – based on a recently published event-based architecture called the EGRU. SpiNNaker2 is a many-core neuromorphic chip designed for large-scale asynchronous processing, and the EGRU is architected to leverage such hardware efficiently while maintaining competitive task performance. This implementation marks the first time a neuromorphic language model matches LSTMs, setting the stage for taking task performance to the level of large language models. We also demonstrate results on a gesture recognition task based on inputs from a DVS camera. Overall, our results showcase the feasibility of this neuro-inspired neural network in hardware, highlighting significant gains versus conventional hardware in energy efficiency for the common use case of single batch inference.

关键词： Performance evaluation Neuromorphics Large language models Neural networks Spiking neural networks Hardware Task analysis

Semantic Encoder Guided Generative Adversarial Face Ultra-Resolution Network

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Wang, Xiang Yang, Yimin Pang, Qixiang Lu, Xiao Liu, Yu Du, Shan Dept. of Computer Science Lakehead University Thunder BayON Canada Dept. of Electrical and Computer Engineering Western University LondonON Canada Okanagan College KelownaBC Canada College of Engineering and Design Hunan Normal University Hunan China School of Microelectronics Tianjin University Tianjin China Dept. of Comp. Sci. Math Physics & Statistics The University of British Columbia OkanaganBC Canada

Face super-resolution is a domain-specific image super-resolution, which aims to generate High-Resolution (HR) face images from their Low-Resolution (LR) counterparts. In this paper, we propose a novel face super-resolution method, namely Semantic Encoder guided Generative Adversarial Face Ultra-Resolution Network (SEGA-FURN) to ultra-resolve an unaligned tiny LR face image to its HR counterpart with multiple ultra-upscaling factors (e.g., 4× and 8×). The proposed network is composed of a novel semantic encoder that has the ability to capture the embedded semantics to guide adversarial learning and a novel generator that uses a hierarchical architecture named Residual in Internal Dense Block (RIDB). Moreover, we propose a joint discriminator which discriminates both image data and embedded semantics. The joint discriminator learns the joint probability distribution of the image space and latent space. We also use a Relativistic average Least Squares loss (RaLS) as the adversarial loss to alleviate the gradient vanishing problem and enhance the stability of the training procedure. Extensive experiments on large face datasets have proved that the proposed method can achieve superior super-resolution results and significantly outperform other state-of-the-art methods in both qualitative and quantitative comparisons. © 2022, CC BY.

关键词： Semantics

FragQC: An Efficient Quantum Error Reduction Technique using Quantum Circuit Fragmentation

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Basu, Saikat Das, Arnav Saha, Amit Chakrabarti, Amlan Sur-Kolay, Susmita A.K.Choudhury School of Information Technology U. of Calcutta Kolkata India LTIMindtree Ltd. Kolkata India Dept. of Computer Science St. Xavier's University Kolkata India Pune India Advanced Computing and Microelectronics Unit Indian Statistical Institute Kolkata India

Quantum computers must meet extremely stringent qualitative and quantitative requirements on their qubits in order to solve real-life problems. Quantum circuit fragmentation techniques divide a large quantum circuit into a number of sub-circuits that can be executed on the smaller noisy quantum hardware available. However, the process of quantum circuit fragmentation involves finding an ideal cut that has exponential time complexity, and also classical post-processing required to reconstruct the output. In this paper, we represent a quantum circuit using a weighted graph and propose a novel classical graph partitioning algorithm for selecting an efficient fragmentation that reduces the entanglement between the sub-circuits along with balancing the estimated error in each sub-circuit. We also demonstrate a comparative study over different classical and quantum approaches of graph partitioning for finding such a cut. We present FragQC, a software tool that cuts a quantum circuit into sub-circuits when its error probability exceeds a certain threshold. With this proposed approach, we achieve an increase of fidelity by 14.83% compared to direct execution without cutting the circuit, and 8.45% over the state-of-the-art ILP-based method, for the benchmark circuits. © 2023, CC BY.

关键词： Genetic algorithms

Dynamic Vth in p-GaN Gate Power HEMTs and Its Impacts upon Power Switching Circuits 15

学校读者我要写书评

暂无评论

Dynamic Vth in p-GaN Gate Power HEMTs and Its Impacts upon P...

15th IEEE International Conference on Solid-State and Integrated Circuit Technology, ICSICT 2020

作者： Wei, Jin Xu, Han Xie, Ruiliang Chen, Kevin J. Institute of Microelectronics Peking University Beijing100871 China Hong Kong University of Science and Technology Dept. of Electronic and Computer Engineering Hong Kong Hong Kong

ISBN: (纸本)9781728162355

The p-GaN gate power HEMT exhibits a dynamic threshold voltage (V-{\mathrm{th}}) during device operation, i.e. its V-{\mathrm{th}} is a mathematical function of the terminal bias. The dynamic V-{\mathrm{th}} phenomenon can be accurately predicted by a charge storage mechanism. The floating p-GaN layer stores negative charges during switching operating, and requires an additional gate bias to counteract these charges. The dynamic V-{\mathrm{th}} has profound impacts upon power switching circuits. It affects the dynamic on-state characteristics and reverse conduction characteristics, presents restriction on the design of gate driving circuit design, and influences the switching waveforms of the circuits. Detailed analysis of these effects will be presented. © 2020 IEEE.

关键词： Timing circuits

Capacitance Deviations for Deformed Parts of Solid MEMS Sensors

学校读者我要写书评

暂无评论

Capacitance Deviations for Deformed Parts of Solid MEMS Sens...

International Conference on Perspective Technologies and Methods in MEMS Design (MEMSTECH)

作者： Jacek Nazdrowicz Adam Stawiński Andrzej Napieralski Dept. of Microelectronics and Computer Science Lodz University of Technology Lodz Poland

In this paper authors presents considerations regarding capacitance deviations under structure deformations. This problem is very important because meaningfully influences on device sensing capability by changes of ou... 详细信息

ISBN: (数字)9781728171807

ISBN: (纸本)9781728171814

关键词： MEMS accelerometer vibratory gyroscope thermal expansion temperature effect capacitance Vibratory gyroscopes capacitance Microelectromechanical systems accelerometers temperature effects THERMAL EXPANSION Capacitors output capacitor electrical signal

Weight Sparsity Complements Activity Sparsity in Neuromorphic Language Models

学校读者我要写书评

暂无评论

Weight Sparsity Complements Activity Sparsity in Neuromorphi...

Neuromorphic Systems (ICONS), International Conference on

作者： Rishav Mukherji Mark Schöne Khaleelulla Khan Nazeer Christian Mayr David Kappel Anand Subramoney Birla Institute of Technology and Science Pilani - Goa Campus Goa India Chair of Highly-Parallel VLSI-Systems and Neuro-Microelectronics TU Dresden Dresden Germany Center for Scalable Data Analytics and Artificial Intelligence (ScaDS.AI) TU Dresden Dresden Germany Centre for Tactile Internet (CeTI) with Human-in-the-Loop TU Dresden Dresden Germany Ruhr University Bochum Institut für Neuroinformatik Germany Dept. of Computer Science Royal Holloway University of London Egham United Kingdom

ISBN: (数字)9798350368659

ISBN: (纸本)9798350368666

Activity and parameter sparsity are two standard methods of making neural networks computationally more efficient. Event-based architectures such as spiking neural networks (SNNs) naturally exhibit activity sparsity, and many methods exist to sparsify their connectivity by pruning weights. While the effect of weight pruning on feed-forward SNNs has been previously studied for computer vision tasks, the effects of pruning for complex sequence tasks like language modeling are less well studied since SNNs have traditionally struggled to achieve meaningful performance on these tasks. Using a recently published SNN-like architecture that works well on small-scale language modeling, we study the effects of weight pruning when combined with activity sparsity. Specifically, we study the tradeoff between the multiplicative efficiency gains the combination affords and its effect on task performance for language modeling. To dissect the effects of the two sparsities, we conduct a comparative analysis between densely activated models and sparsely activated event-based models across varying degrees of connectivity sparsity. We demonstrate that sparse activity and sparse connectivity complement each other without a proportional drop in task performance for an event-based neural network trained on the Penn Treebank and WikiText-2 language modeling datasets. Our results suggest sparsely connected event-based neural networks are promising candidates for effective and efficient sequence modeling.

关键词： Analytical models computer vision Event detection Neuromorphics Computational modeling Neural networks computer architecture Spiking neural networks Computational efficiency Standards