2021/03/01 by Yi Wei Daniel Tay, Tay, Yi, Dehghani, Mostafa +13 · 4 citations
Computer Science · #Advanced Neural Network Applications #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis #Machine Learning (cs.LG)
paper · pdf · doi:10.48550/arxiv.2103.01075
openalex publication_date 2021/03/01 · openalex created_date 2021/03/15 · openalex updated_date 2026/07/28
This paper proposes Omnidirectional Representations from Transformers (OmniNet). In OmniNet, instead of maintaining a strictly horizontal receptive field, each token is allowed to attend to all tokens in the entire network. This process can also be interpreted as a form of extreme or intensive attention mechanism that has the receptive field of the entire width and depth of the network. To this end, the omnidirectional attention is learned via a meta-learner, which is essentially another self-attention based model. In order to mitigate the computationally expensive costs of full receptive field attention, we leverage efficient self-attention models such as kernel-based (Choromanski et al.), low-rank attention (Wang et al.) and/or Big Bird (Zaheer et al.) as the meta-learner. Extensive experiments are conducted on autoregressive language modeling (LM1B, C4), Machine Translation, Long Range Arena (LRA), and Image Recognition. The experiments show that OmniNet achieves considerable improvements across these tasks, including achieving state-of-the-art performance on LM1B, WMT'14 En-De/En-Fr, and Long Range Arena. Moreover, using omnidirectional representation in Vision Transformers leads to significant improvements on image recognition tasks on both few-shot learning and fine-tuning setups.