Comparing Attention-based Convolutional and Recurrent Neural Networks: Success and Limitations in Machine Reading Comprehension
We propose a machine reading comprehension model based on the compare-aggregate framework with two-staged attention that achieves state-of-the-art results on the MovieQA question answering dataset. To investigate the limitations of our model as well as the behavioral difference between convolutional and recurrent neural networks, we generate adversarial examples to confuse the model and compare to human performance. Furthermore, we assess the generalizability of our model by analyzing its differences to human inference,
Code (1)
Tasks
Machine Reading ComprehensionQuestion AnsweringReading ComprehensionSimilar Papers 제목 키워드 기반
A Survey of Vision Transformers in Autonomous Driving: Current Trends and Future Directions
This survey explores the adaptation of visual transformer models in Autonomous Driving, a transition inspired by their success in Natural Language Processing. Surpassing traditional Recurrent Neural Networks in tasks lik…
Autonomous DrivingDecoderLane Detectionobject-detection+4CRAM: Clued Recurrent Attention Model
To overcome the poor scalability of convolutional neural network, recurrent attention model(RAM) selectively choose what and where to look on the image. By directing recurrent attention model how to look the image, RAM c…
Decoderimage-classificationImage ClassificationmodelAnalysis of Convolutional Decoder for Image Caption Generation
Recently Convolutional Neural Networks have been proposed for Sequence Modelling tasks such as Image Caption Generation. However, unlike Recurrent Neural Networks, the performance of Convolutional Neural Networks as Deco…
Caption GenerationData AugmentationDecoderImage Captioning+1Convolutional Recurrent Neural Network with Attention for 3D Speech Enhancement
3D speech enhancement can effectively improve the auditory experience and plays a crucial role in augmented reality technology. However, traditional convolutional-based speech enhancement methods have limitations in extr…
DenoisingSpeech EnhancementTransformer based Grapheme-to-Phoneme Conversion
Attention mechanism is one of the most successful techniques in deep learning based Natural Language Processing (NLP). The transformer network architecture is completely based on attention mechanisms, and it outperforms …
Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Grapheme-to-Phoneme ConversionMachine Translation+5