Techniques

CvT: Introducing Convolutions to Vision Transformers ICCV 2021 CWY Parametrization: a Solution for Parallelized Optimization of Orthogonal and Stiefel Matrices AISTATS 2021 DANCE: A Deep Attentive Contour Model for Efficient Instance Segmentation WACV 2021 DAP: Detection-Aware Pre-Training With Weak Supervision CVPR 2021 DARTS-: Robustly Stepping out of Performance Collapse Without Indicators ICLR 2021 Data Augmentation for Voice-Assistant NLU using BERT-based Interchangeable Rephrase EACL 2021 Data-Efficient Alignment of Multimodal Sequences by Aligning Gradient Updates and Internal Feature Distributions WACV 2021 Data-Efficient Language Shaped Few-shot Image Classification EMNLP 2021 DBNet: A Dual-Branch Network Architecture Processing on Spectrum and Waveform for Single-Channel Speech Enhancement Interspeech 2021 DDSAS: Dynamic and Differentiable Space-Architecture Search ACML 2021 DEBERTA: DECODING-ENHANCED BERT WITH DISENTANGLED ATTENTION ICLR 2021 DebiNet: Debiasing Linear Models with Nonlinear Overparameterized Neural Networks AISTATS 2021 Deconstructing the Regularization of BatchNorm ICLR 2021 Decoupled Dynamic Filter Networks CVPR 2021 Deep Analysis of CNN-Based Spatio-Temporal Representations for Action Recognition CVPR 2021 Deep Blind Video Super-Resolution ICCV 2021 Deep Frequency Principle Towards Understanding Why Deeper Learning Is Faster AAAI 2021 Deep Gradient Projection Networks for Pan-sharpening CVPR 2021 Deep Learning for Functional Data Analysis with Adaptive Basis Layers ICML 2021 Deep Learning Through the Lens of Example Difficulty NeurIPS 2021 DeepLM: Large-Scale Nonlinear Least Squares on Deep Learning Frameworks Using Stochastic Domain Decomposition CVPR 2021 Deep Low-Contrast Image Enhancement using Structure Tensor Representation AAAI 2021 Deeply Shape-Guided Cascade for Instance Segmentation CVPR 2021 Deep Matching Prior: Test-Time Optimization for Dense Correspondence ICCV 2021 Deep Neural Network Calibration for E2E Speech Recognition System Interspeech 2021