检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Wang, Kesu Yan, Meng Zhang, He Hu, Haibo State Key Lab for Novel Software Technology Nanjing University Nanjing China School of Big Data & Software Engineering Chongqing University Chongqing China

Program classification can be regarded as a high-level abstraction of code, laying a foundation for various tasks related to source code comprehension, and has a very wide range of applications in the field of software engineering, such as code clone detection, code smell classification, defects classification, etc. The cross-language program classification can realize code transfer in different programming languages, and can also promote cross-language code reuse, thereby helping developers to write code quickly and reduce the development time of code transfer. Most of the existing studies focus on the semantic learning of the code, whilst few studies are devoted to cross-language tasks. The main challenge of cross-language program classification is how to extract semantic features of different programming languages. In order to cope with this difficulty, we propose a Unified Abstract Syntax Tree (namely UAST in this paper) neural network. In detail, the core idea of UAST consists of two unified mechanisms. First, UAST learns an AST representation by unifying the AST traversal sequence and graph-like AST structure for capturing semantic code features. Second, we construct a mechanism called unified vocabulary, which can reduce the feature gap between different programming languages, so it can achieve the role of cross-language program classification. Besides, we collect a dataset containing 20,000 files of five programming languages, which can be used as a benchmark dataset for the cross-language program classification task. We have done experiments on two datasets, and the results show that our proposed approach outperforms the state-of-the-art baselines in terms of four evaluation metrics (Precision, Recall, F1-score, and Accuracy). Copyright © 2022, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

DomainAdaptor: A Novel Approach to Test-time Adaptation

arXiv

引用

arXiv 2023年

作者： Zhang, Jian Qi, Lei Shi, Yinghuan Gao, Yang State Key Laboratory for Novel Software Technology Nanjing University China National Institute of Healthcare Data Science Nanjing University China School of Computer Science and Engineering Southeast University China

To deal with the domain shift between training and test samples, current methods have primarily focused on learning generalizable features during training and ignore the specificity of unseen samples that are also critical during the test. In this paper, we investigate a more challenging task that aims to adapt a trained CNN model to unseen domains during the test. To maximumly mine the information in the test data, we propose a unified method called DomainAdaptor for the test-time adaptation, which consists of an AdaMixBN module and a Generalized Entropy Minimization (GEM) loss. Specifically, AdaMixBN addresses the domain shift by adaptively fusing training and test statistics in the normalization layer via a dynamic mixture coefficient and a statistic transformation operation. To further enhance the adaptation ability of AdaMixBN, we design a GEM loss that extends the Entropy Minimization loss to better exploit the information in the test data. Extensive experiments show that DomainAdaptor consistently outperforms the state-of-the-art methods on four benchmarks. Furthermore, our method brings more remarkable improvement against existing methods on the few-data unseen domain. The code is available at https://***/koncle/DomainAdaptor. Copyright © 2023, The Authors. All rights reserved.

关键词： Entropy

来源：评论

学校读者我要写书评

暂无评论

Nano-wire bridge waveguide-assisted silicon polarizing beam splitter 13

Nano-wire bridge waveguide-assisted silicon polarizing beam ...

引用

Holography, Diffractive Optics, and Applications XIII 2023

作者： Liu, H.P. Feng, J.J. Chen, J. Zhou, W.J. Bi, Q.Y. Zeng, H.P. School of Optical-Electrical and Computer Engineering University of Shanghai for Science and Technology Shanghai200093 China Key Lab of Advanced Optical Manufacturing Technologies of Jiangsu Province Soochow University Suzhou215006 China Chongqing Key Laboratory of Precision Optics Chongqing Institute of East China Normal University Chongqing401120 China State Key Laboratory of Precision Spectroscopy East China Normal University Shanghai200241 China

ISBN: (纸本)9781510667853

A low-loss and wideband silicon polarizing beam splitter is demonstrated with the assistance of a nano-bridge waveguide. Transverse magnetic light can be coupled to the cross port through the nano-bridge waveguide, while the transverse electric light comes out mainly from the through port. The designed device has a coupling length of 19.6 µm, which can realize an extinction ratio of 32.43 dB for TM mode or 34.23 dB for TE mode at a 1565-nm wavelength. Compared with the conventional three-waveguide coupler structure, the proposed device is based on a resonant tunneling principle, which can help to effectively improve the fabrication tolerance. The device is fabricated with a commercial CMOS processing facility, which can achieve an extinction ratio of 23.07 dB for TM or 23.46 dB for TE mode with a low excess loss, and the extinction ratio of more than 10 dB can be realized in the wavelength range from 1525 to 1610 nm for both modes. The device performance can be further improved, which would facilitate its practical applications in commercial integrated optical circuits. © 2023 SPIE.

关键词： Silicon photonics

来源：评论

学校读者我要写书评

暂无评论

SLAM: Towards Efficient Multilingual Reasoning via Selective Language Alignment

arXiv

引用

arXiv 2025年

作者： Fan, Yuchun Mu, Yongyu Wang, Yilin Huang, Lei Ruan, Junhao Li, Bei Xiao, Tong Huang, Shujian Feng, Xiaocheng Zhu, Jingbo NLP Lab School of Computer Science and Engineering Northeastern University Shenyang China NiuTrans Research Shenyang China Harbin Institute of Technology Harbin China Meituan Inc China National Key Laboratory for Novel Software Technology Nanjing University China

Despite the significant improvements achieved by large language models (LLMs) in English reasoning tasks, these models continue to struggle with multilingual reasoning. Recent studies leverage a full-parameter and two-stage training paradigm to teach models to first understand non-English questions and then reason. However, this method suffers from both substantial computational resource computing and catastrophic forgetting. The fundamental cause is that, with the primary goal of enhancing multilingual comprehension, an excessive number of irrelevant layers and parameters are tuned during the first stage. Given our findings that the representation learning of languages is merely conducted in lower-level layers, we propose an efficient multilingual reasoning alignment approach that precisely identifies and finetunes the layers responsible for handling multilingualism. Experimental results show that our method, SLAM, only tunes 6 layers' feedforward sub-layers including 6.5-8% of all parameters within 7B and 13B LLMs, achieving superior average performance than all strong baselines across 10 languages. Meanwhile, SLAM only involves one training stage, reducing training time by 4.1 - 11.9× compared to the two-stage method. Copyright © 2025, The Authors. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Sphinx: A Hybrid Boolean Processor-FPGA Hardware Emulation System

Sphinx: A Hybrid Boolean Processor-FPGA Hardware Emulation S...

引用

IEEE International Conference on computer-Aided Design

作者： Ruiyao Pu Yiwei Sun Pei-Hsin Ho Fan Yang Li Shang Xuan Zeng State Key Lab of Integrated Chips and Systems School of Microelectronics Fudan University Shanghai China UniVista Industrial Software Group Shanghai China China and Shanghai Key Laboratory of Data Science School of Computer Science Fudan University Shanghai China

Existing hardware emulators use either FPGA or Boolean processors, which suffer from long compile time and poor debuggability (FPGA-based), or low emulation performance (Boolean processor-based). This work presents Sphinx, a hybrid Boolean processor-FPGA hardware emulation platform aiming to overcome these shortcomings. Sphinx hardware is a new hybrid architecture that integrates software programmable Boolean processors and FPGAs. Sphinx software is a compilation framework that conducts incremental design partitioning and implements the design-under-test components on Boolean processors and the rest on FPGAs. Together, Sphinx enables an incremental emulation flow and demonstrates high emulation performance, fast compile turnarounds, and good debuggability.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Detecting Memory-Related Bugs by Tracking Heap Memory Management of C++ Smart Pointers 36

Detecting Memory-Related Bugs by Tracking Heap Memory Manage...

引用

36th IEEE/ACM International Conference on Automated software engineering, ASE 2021

作者： Ma, Xutong Yan, Jiwei Wang, Wei Yan, Jun Zhang, Jian Qiu, Zongyan Chinese Academy of Sciences State Key Laboratory of Computer Science Institute of Software China Technology Center of Software Engineering Institute of Software Chinese Academy of Sciences China University of Chinese Academy of Sciences China Peking University School of Mathematical Sciences China

ISBN: (纸本)9781665403375

The smart pointer mechanism, which is improved in the continuous versions of the C++ standards over the last decade, is designed to prevent memory-leak bugs by automatically deallocating the managed memory blocks. However, not all kinds of memory errors can be immunized by adopting this mechanism. For example, dereferencing a null smart pointer will lead to a software failure. Due to the lack of specialized support for smart pointers, the off-the-shelf C++ static analyzers cannot effectively reveal these *** this paper, we propose a static approach to detecting memory-related bugs by tracking the heap memory management of smart pointers. The behaviors of smart pointers are modeled during their lifetime to trace the state transitions of managed memory blocks. and the specially designed checkers are used to check the state changes according to five collected bug patterns. To evaluate the effectiveness of our approach, we implement it on the top of the Clang Static Analyzer. A set of handmade code snippets, as well as nine popular open-source C++ projects, are used to compare our tool against four other analyzers. The results show that our approach can successfully discover nearly all the built-in bugs. and 442 out of 648 reports generated from the open-source projects are true positives after manual reviewing, where the bugs of dereferencing null smart pointers are most frequently reported. To further confirm our reports, we design patches for Aria2, Restbed, MySQL and LLVM, in which seven pull requests covering 76 bug reports have been merged by the developers up to now. The results indicate that pointers should always be carefully used even after migrated to smart pointers and static analysis upon specialized models can effectively detect such bugs. © 2021 IEEE.

关键词： Static analysis

来源：评论

学校读者我要写书评

暂无评论

Transferable Waveform-level Adversarial Attack against Speech Anti-spoofing Models

Transferable Waveform-level Adversarial Attack against Speec...

引用

IEEE International Conference on Multimedia and Expo (ICME)

作者： Bingyuan Huang Sanshuai Cui Xiangui Kang Enping Li Guangdong Key Lab of Information Security School of Computer Science and Engineering Sun Yat-Sen University Guangzhou China Computer Science Department Bridgewater State University Massachusetts USA

Speech anti-spoofing models protect media from malicious fake speech but are vulnerable to adversarial attacks. Studies of adversarial attacks are conducive to developing robust speech anti-spoofing systems. Existing transfer-based attack methods mainly craft adversarial speech examples at the handcrafted-feature level, which have limited attack ability against the real-world anti-spoofing systems, as these systems only have raw waveform input interfaces. In this work, we propose a waveform-level input data transformation, called the temporal smoothing method, to generate more transferable adversarial speech examples. In the optimization iterations of the adversarial perturbation, we randomly smooth input waveforms to prevent the adversarial examples from overfitting white-box surrogate models. The proposed transformation can be combined with any iterative gradient-based attack method. Extensive experiments demonstrate that our method significantly enhances the transferability of waveform-level adversarial speech examples.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Outlier Suppression+: Accurate quantization of large language models by equivalent and optimal shifting and scaling

arXiv

引用

arXiv 2023年

作者： Wei, Xiuying Zhang, Yunchen Li, Yuhang Zhang, Xiangguo Gong, Ruihao Guo, Jinyang Liu, Xianglong State Key Lab of Software Development Environment Beihang University China SenseTime Research China School of Computer and Communication Sciences EPFL Switzerland Yale University United States UESTC China

Post-training quantization (PTQ) of transformer language models faces significant challenges due to the existence of detrimental outliers in activations. We observe that these outliers are concentrated in specific channels and are asymmetric across channels. To address this issue, we propose the Outlier Suppression+ (OS+) framework, which contains the channel-wise shifting for asymmetry and channel-wise scaling for concentration. We show that these operations can be seamlessly migrated into subsequent modules while maintaining equivalence. Second, we propose a fast and stable scheme to calculate effective shifting and scaling values. The channel-wise shifting aligns the center of each channel for removal of outlier asymmetry. The channel-wise scaling quantitatively evaluates changes brought by migration and quantization for better quantization burden balance. We validate our OS+ under both standard and fine-grained quantization settings with models including BERT, OPT, BLOOM, BLOOMZ, and LLaMA. Comprehensive results across various tasks demonstrate the superiority of our approach. Especially, with standard quantization, OS+ can achieve near-floating-point performance on both small models and large language models on 8-bit and 6-bit. Besides, we establish a new state-of-the-art for 4-bit BERT with 15.5% improvement. Our code is available at https://***/ModelTC/Outlier_Suppression_Plus. Copyright © 2023, The Authors. All rights reserved.

关键词： Digital arithmetic

来源：评论

学校读者我要写书评

暂无评论

Explaining Model Overfitting in CNNS Via GMM Clustering

SSRN

引用

SSRN 2024年

作者： Dou, Hui Mu, Xinyu Yi, Mengjun Han, Feng Zhao, Jian Shen, Furao State Key Laboratory for Novel Software Technology China Department of Computer Science and Technology Nanjing University Nanjing210023 China School of Electronic Science and Engineering Nanjing University Nanjing210023 China School of Artificial Intelligence Nanjing University Nanjing210023 China

Convolutional Neural Networks (CNNs) have demonstrated remarkable prowess in the field of computer vision. However, their opaque decision-making processes pose significant challenges for practical applications. In this study, we provide quantitative metrics for assessing CNN filters by clustering the feature maps corresponding to individual filters in the model via Gaussian Mixture Model (GMM). By analyzing the clustering results, we screen out some anomaly filters associated with outlier samples. We further analyze the relationship between the anomaly filters and model overfitting, proposing three hypotheses. This method is universally applicable across diverse CNN architectures without modifications, as evidenced by its successful application to models like AlexNet and LeNet-5. We present three meticulously designed experiments demonstrating our hypotheses from the perspectives of model behavior, dataset characteristics, and filter impacts. Through this work, we offer a novel perspective for evaluating the CNN performance and gain new insights into the operational behavior of model overfitting. © 2024, The Authors. All rights reserved.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Explaining Model Overfitting in CNNs via GMM Clustering

arXiv

引用

arXiv 2024年

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：