Commit
•
efb1f2a
1
Parent(s):
85ba1ba
encoder-encoder -> encoder-decoder
Browse filesIn the [paper](https://arxiv.org/abs/2107.07653):
`TA PE X is conceptually simple and easy to implement. In this paper, we regard the pre-training as a sequence generation task and employ an encoder-decoder model`
README.md
CHANGED
@@ -14,7 +14,7 @@ TAPEX was proposed in [TAPEX: Table Pre-training via Learning a Neural SQL Execu
|
|
14 |
|
15 |
TAPEX (**Ta**ble **P**re-training via **Ex**ecution) is a conceptually simple and empirically powerful pre-training approach to empower existing models with *table reasoning* skills. TAPEX realizes table pre-training by learning a neural SQL executor over a synthetic corpus, which is obtained by automatically synthesizing executable SQL queries.
|
16 |
|
17 |
-
TAPEX is based on the BART architecture, the transformer encoder-
|
18 |
|
19 |
## Intended Uses
|
20 |
|
|
|
14 |
|
15 |
TAPEX (**Ta**ble **P**re-training via **Ex**ecution) is a conceptually simple and empirically powerful pre-training approach to empower existing models with *table reasoning* skills. TAPEX realizes table pre-training by learning a neural SQL executor over a synthetic corpus, which is obtained by automatically synthesizing executable SQL queries.
|
16 |
|
17 |
+
TAPEX is based on the BART architecture, the transformer encoder-decoder (seq2seq) model with a bidirectional (BERT-like) encoder and an autoregressive (GPT-like) decoder.
|
18 |
|
19 |
## Intended Uses
|
20 |
|