Update README.md
Browse files
README.md
CHANGED
|
@@ -6,7 +6,7 @@ training an intent classifier or a slot tagger, for example, we can ask the mode
|
|
| 6 |
slot-related questions in natural language:
|
| 7 |
|
| 8 |
```
|
| 9 |
-
Context : I'm looking for a cheap flight to Boston.
|
| 10 |
|
| 11 |
Question: Is the user looking to book a flight?
|
| 12 |
Answer : Yes
|
|
@@ -21,6 +21,8 @@ Question: Where is the user flying from?
|
|
| 21 |
Answer : (empty)
|
| 22 |
```
|
| 23 |
|
|
|
|
|
|
|
| 24 |
Thus, by asking questions for each intent and slot in natural language, we can effectively construct an NLU hypothesis. For more details, please read the paper: [Language model is all you need: Natural language understanding as question answering](https://assets.amazon.science/33/ea/800419b24a09876601d8ab99bfb9/language-model-is-all-you-need-natural-language-understanding-as-question-answering.pdf).
|
| 25 |
|
| 26 |
## Model training
|
|
|
|
| 6 |
slot-related questions in natural language:
|
| 7 |
|
| 8 |
```
|
| 9 |
+
Context : Yes. No. I'm looking for a cheap flight to Boston.
|
| 10 |
|
| 11 |
Question: Is the user looking to book a flight?
|
| 12 |
Answer : Yes
|
|
|
|
| 21 |
Answer : (empty)
|
| 22 |
```
|
| 23 |
|
| 24 |
+
Note the "Yes. No. " prepended in the context. Those are to allow the model to answer intent-related questions (`Is the user looking for a restaurant?`).
|
| 25 |
+
|
| 26 |
Thus, by asking questions for each intent and slot in natural language, we can effectively construct an NLU hypothesis. For more details, please read the paper: [Language model is all you need: Natural language understanding as question answering](https://assets.amazon.science/33/ea/800419b24a09876601d8ab99bfb9/language-model-is-all-you-need-natural-language-understanding-as-question-answering.pdf).
|
| 27 |
|
| 28 |
## Model training
|