- Activation Maximization Generative Adversarial Nets (민국)
- Top-k Training of GANs: Improving GAN Performance by Throwing Away Bad Samples (민국)
- Lipschitz Generative Adversarial Nets (민국)
- Closed-Form Factorization of Latent Semantics in GANs (민국)
- Your GAN is secretly an Energy-based Model and You Should Use Discriminator Driven Latent Sampling (민국)
- Transpose: Keypoint Localization via Transformer (승욱)
- RoFormer: Enhanced Transformer with Rotary Position Embedding (승욱)
- Recurrent Parameter Generators (승욱)
- Harmonic Networks: Deep Translation and Rotation Equivariance (우현)
- Deep Learning on Lie Groups for Skeleton-based Action Recognition (우현)
- Infinite Memory Transformers (승욱)
- Sparse-MLP: A Fully-MLP Architecture with Conditional Computation (승욱)
- An Image is worth 16x16 words : Transformers for Image Recognition at Scale(수현)
- End to End object detection with Transformers (수현)
- Eyes Tell All: Irregular Pupil Shapes Reveal GAN-Generated Faces (승욱)
- Swapping Autoencoder for Deep Image Manipulation (진오)
- InfoGraph: Unsupervised and semi-supervised graph-level representation learning via mutual infromation maximization (지예)
- On the bottleneck of graph neural networks and its practical implications (지예)
- Constrained Generation of Semantically Valid graph via regularizing variational autoencoders (지예)
- How to find your friendly neighborhood: Graph attention design with self-supervision (지예)
- ASMR: Learning Attribute-Based Person Search with Adaptive Semantic Margin Regularizer (지예)
- Using rule-based labels for weak supervised learning: A chemnet for transferable chemical property prediction (지예)
- Molecule attention transformer (지예)
- Speaker sensitive response evaluation model (지예)
- Junction Tree variational autoencoder for molecular graph generation (지예)
- On the evaluation of conditional generation (FJD) (지예)
- ConvMLP: Hierarchical Convolutional MLPs for Vision (승욱)
- Movement Pruning: Adaptive Sparsity by Fine-Tuning (승욱)
- MyTraces: Investigating Correlation and Causation between User's Emotional States and Mobile Phone Interaction (승욱)
- SonicPrint: A Generally adoptable and secure fingerprint biometrics in Smart Devices (승욱)
- torch.manual seed(3407) is all you need: On the influence of random seeds in deep learning architectures for computer vision(승욱)
- Dynamic Convolution: Attention over Convolution Kernels (Yoonwoo)
- Involution: Inverting the Inherence of Convolutiojns for Visual Recognition (Yoonwoo)
- PAConv: Position Adaptive Convolution with Dynamic Kernel Assembling on Point Clouds (Yoonwoo)
- PVT: Point Voxel Transformer for 3D Deep Learning (Yoonwoo)
- Voxel Transformer for 3D Object Detection (Yoonwoo)