什么?" 万能的 "Transformer即将被踢下神坛...
共 4039字,需浏览 9分钟
·
2021-02-12 12:55
谷歌研究科学家 David Ha:Transformer 是新的 LSTM。
论文:Taming Transformers for High-Resolution Image Synthesis
链接:https://arxiv.org/pdf/2012.09841v1.pdf
论文:TransTrack: Multiple-Object Tracking with Transformer
链接:https://arxiv.org/pdf/2012.15460v1.pdf
论文:Compound Word Transformer: Learning to Compose Full-Song Music over Dynamic Directed Hypergraphs
链接:https://arxiv.org/pdf/2101.02402v1.pdf
论文:Dance Revolution: Long-Term Dance Generation with Music via Curriculum Learning
链接:https://arxiv.org/pdf/2006.06119v5.pdf
论文:Self-Attention Based Context-Aware 3D Object Detection
链接:https://arxiv.org/pdf/2101.02672v1.pdf
论文:PCT: Point Cloud Transformer
链接:https://arxiv.org/pdf/2012.09688v1.pdf
论文:Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecasting
链接:https://arxiv.org/pdf/1912.09363v3.pdf
论文:VinVL: Making Visual Representations Matter in Vision-Language Models
链接:https://arxiv.org/pdf/2101.00529v1.pdf
论文:End-to-end Lane Shape Prediction with Transformers
链接:https://arxiv.org/pdf/2011.04233v2.pdf
论文:Deformable DETR: Deformable Transformers for End-to-End Object Detection
链接:https://arxiv.org/pdf/2010.04159v2.pdf
本文部分素材来源于网络,如有侵权,联系删除。
实“鼠”不易 的2020年过去了,在大年三十儿晚上,真诚地祝大家新年快乐!
新的一年,七月在线将努力输出更多精品干货!最后,再次祝大家新年快乐,牛气冲天,牛牛牛~
特此,为大家准备了年度福利——特训好课 0.01元 秒杀