AttnFuse: A Composable DSL for Compiling Attentions to Fused GPU Kernels
为解决Transformer中注意力机制的计算和内存成本问题,提出AttnFuse DSL,通过编译成融合GPU内核来提高执行效率。
为解决Transformer中注意力机制的计算和内存成本问题,提出AttnFuse DSL,通过编译成融合GPU内核来提高执行效率。
研究通过数据驱动框架解决对话中参照不确定性下的选择性信念修正问题,对比不同策略后发现累积不确定性而非局部差异更促进有效理解。
本文通过利用样本分位数的联合正态性和delta方法,提出四种稳健估计器(oQLS、gQLS、log-oQLS和log-gQLS),以解决对数位置尺度损失模型的拟合与验证问题。
This study investigates the differential effects of conversational and search-based tools on informal learning regarding controversial topics. Through a crowdsourced comparative experiment, we analyzed how user characteristics and interaction patterns influence learning outcomes and critical reflection. Results indicate no significant differences between the two tools in terms of learning gains or critical reflection. Instead, individual traits such as attitude strength and intellectual humility were found to be stronger determinants of immediate learning efficacy than the specific information retrieval modality employed. These findings suggest that in the context of controversial subject matter, individual cognitive attributes exert a more pivotal influence than technological form factors. Consequently, this work offers novel insights into learning mechanisms within human-computer interaction by highlighting the primacy of learner dispositions over tool design in shaping educational outcomes for contentious issues.
This study addresses the lack of systematic comparison among convolutional neural networks (CNNs), vision Transformers (ViTs), and hybrid architectures for colorectal histopathological image classification. Leveraging the Kather dataset, the authors evaluate twelve prominent models—including ResNet34, ConvNeXt-Tiny, ViT-B/16, and EVA-02—under a unified pipeline of ImageNet pretraining, transfer learning, and fine-tuning. For the first time, the three major deep learning paradigms are comprehensively benchmarked under standardized conditions, with performance analyzed through accuracy, interpretability, and per-class metrics to identify strengths and challenging categories. Results show that all models achieve high accuracy ranging from 93.2% to 97.1%, with the hybrid model EVA-02 attaining the best performance at 97.1%. Vision Transformers generally outperform CNNs, while modern CNNs offer a favorable trade-off between accuracy and computational complexity.
为解决Transformer中注意力机制的计算和内存成本问题,提出AttnFuse DSL,通过编译成融合GPU内核来提高执行效率。
研究通过数据驱动框架解决对话中参照不确定性下的选择性信念修正问题,对比不同策略后发现累积不确定性而非局部差异更促进有效理解。
本文通过利用样本分位数的联合正态性和delta方法,提出四种稳健估计器(oQLS、gQLS、log-oQLS和log-gQLS),以解决对数位置尺度损失模型的拟合与验证问题。
This study investigates the differential effects of conversational and search-based tools on informal learning regarding controversial topics. Through a crowdsourced comparative experiment, we analyzed how user characteristics and interaction patterns influence learning outcomes and critical reflection. Results indicate no significant differences between the two tools in terms of learning gains or critical reflection. Instead, individual traits such as attitude strength and intellectual humility were found to be stronger determinants of immediate learning efficacy than the specific information retrieval modality employed. These findings suggest that in the context of controversial subject matter, individual cognitive attributes exert a more pivotal influence than technological form factors. Consequently, this work offers novel insights into learning mechanisms within human-computer interaction by highlighting the primacy of learner dispositions over tool design in shaping educational outcomes for contentious issues.
This study addresses the lack of systematic comparison among convolutional neural networks (CNNs), vision Transformers (ViTs), and hybrid architectures for colorectal histopathological image classification. Leveraging the Kather dataset, the authors evaluate twelve prominent models—including ResNet34, ConvNeXt-Tiny, ViT-B/16, and EVA-02—under a unified pipeline of ImageNet pretraining, transfer learning, and fine-tuning. For the first time, the three major deep learning paradigms are comprehensively benchmarked under standardized conditions, with performance analyzed through accuracy, interpretability, and per-class metrics to identify strengths and challenging categories. Results show that all models achieve high accuracy ranging from 93.2% to 97.1%, with the hybrid model EVA-02 attaining the best performance at 97.1%. Vision Transformers generally outperform CNNs, while modern CNNs offer a favorable trade-off between accuracy and computational complexity.