timeline
title Evolution of Attention Models
2014 : Seq2Seq (Cho et al. 2014, Sutskever et al. 2014)
2015 : Align & Translate (Bahdanau et al. 2015, Luong et al. 2015)
2015 : Visual attention (Xu et al. 2015)
2016 : Hierarchical attention (Yang et al. 2016)
2017 : Self attention (Vaswani et al. 2017)
2018 : BERT (Devlin et al. 2018)
timeline
title Evolution of GPT and Attention Models
2014 : Seq2Seq (Cho et al. 2014, Sutskever et al. 2014)
2015 : Align & Translate (Bahdanau et al. 2015, Luong et al. 2015)
2015 : Visual attention (Xu et al. 2015)
2016 : Hierarchical attention (Yang et al. 2016)
2017 : Transformer (Vaswani et al. 2017)
2018 : BERT (Devlin et al. 2018)
2018 : GPT (Radford et al. 2018)
2019 : GPT-2 (Radford et al. 2019)
2020 : GPT-3 (Brown et al. 2020)
2022 : GPT-3.5 (OpenAI 2022)
2023 : GPT-4 (OpenAI 2023)