nlp:lstm
Differences
This shows you the differences between two versions of the page.
| Both sides previous revisionPrevious revisionNext revision | Previous revision | ||
| nlp:lstm [2021/02/18 13:38] – jmflanig | nlp:lstm [2023/06/15 07:36] (current) – external edit 127.0.0.1 | ||
|---|---|---|---|
| Line 8: | Line 8: | ||
| < | < | ||
| - | Conneau et al. (2017) explore multiple different sentence embedding architectures | + | [[https:// |
| ranging from LSTM, BiLSTM and intra-attention to convolution neural networks | ranging from LSTM, BiLSTM and intra-attention to convolution neural networks | ||
| and the performance of these architectures on NLI tasks. They show that, out of | and the performance of these architectures on NLI tasks. They show that, out of | ||
| Line 15: | Line 15: | ||
| performance on various transfer learning tasks. | performance on various transfer learning tasks. | ||
| </ | </ | ||
| + | |||
| + | With tweaks, they can outperform Transformer models. | ||
| ===== Resources ===== | ===== Resources ===== | ||
| * [[https:// | * [[https:// | ||
nlp/lstm.1613655527.txt.gz · Last modified: 2023/06/15 07:36 (external edit)