RPyForth: Exposing a Call-Shared Data Stack to a Meta-Tracing JIT Compiler
本文提出了一种固定宽度窗口方法，结合共享溢出区和解码函数，以解决Forth语言在元跟踪JIT编译器中数据栈访问的问题，显著提高了程序执行速度。
0 citationsRead paper
Selected work
Representative Papers
Parameter Estimation of Ringdown Quasinormal Modes with Autoencoder
本文使用自编码器框架解决引力波信号中准正模式参数估计问题，通过训练潜在空间表示单个模式的物理参数，实现波形降噪和参数估计。
0 citationsRead paper
Enhancing Privacy in Federated Learning via Dual Obfuscation of Gradients and Training Images
本文提出一种双重混淆方法，通过同时混淆梯度信息和训练图像来增强联邦学习中的隐私保护，减少数据恢复攻击的风险。
0 citationsRead paper
Privacy-Preserving Object Detection for Vision Transformer-Based Models
本文提出了一种基于感知加密的隐私保护方法，用于视觉变换器模型的对象检测任务，同时保持高精度。
0 citationsRead paper
Declining Modularity of Intellectual Bases During the Emergence of Research Areas
This study addresses the unclear mechanisms underlying knowledge base restructuring during research field formation by proposing a novel framework that tracks modularity decline in co-citation networks. Establishing this decline as a structural signature of integrative field emergence, the research validates the approach through co-citation network analysis, temporal tracking, and statistical robustness assessments. Empirical results demonstrate that this metric effectively identifies emerging or transitional phases across three distinct domains. By elucidating the evolutionary dynamics of cross-community knowledge integration, this work provides a quantifiable structural basis for formulating research strategies and assessing frontier trends, thereby offering critical insights into the structural reconfiguration of scientific knowledge during periods of disciplinary convergence and transformation.
0 citationsRead paper
Recent publications
Latest Papers
RPyForth: Exposing a Call-Shared Data Stack to a Meta-Tracing JIT Compiler
本文提出了一种固定宽度窗口方法，结合共享溢出区和解码函数，以解决Forth语言在元跟踪JIT编译器中数据栈访问的问题，显著提高了程序执行速度。
0 citationsRead paper
Parameter Estimation of Ringdown Quasinormal Modes with Autoencoder
本文使用自编码器框架解决引力波信号中准正模式参数估计问题，通过训练潜在空间表示单个模式的物理参数，实现波形降噪和参数估计。
0 citationsRead paper
Enhancing Privacy in Federated Learning via Dual Obfuscation of Gradients and Training Images
本文提出一种双重混淆方法，通过同时混淆梯度信息和训练图像来增强联邦学习中的隐私保护，减少数据恢复攻击的风险。
0 citationsRead paper
Privacy-Preserving Object Detection for Vision Transformer-Based Models
本文提出了一种基于感知加密的隐私保护方法，用于视觉变换器模型的对象检测任务，同时保持高精度。
0 citationsRead paper
Declining Modularity of Intellectual Bases During the Emergence of Research Areas
This study addresses the unclear mechanisms underlying knowledge base restructuring during research field formation by proposing a novel framework that tracks modularity decline in co-citation networks. Establishing this decline as a structural signature of integrative field emergence, the research validates the approach through co-citation network analysis, temporal tracking, and statistical robustness assessments. Empirical results demonstrate that this metric effectively identifies emerging or transitional phases across three distinct domains. By elucidating the evolutionary dynamics of cross-community knowledge integration, this work provides a quantifiable structural basis for formulating research strategies and assessing frontier trends, thereby offering critical insights into the structural reconfiguration of scientific knowledge during periods of disciplinary convergence and transformation.
0 citationsRead paper