Update README.md
Browse files
README.md
CHANGED
|
@@ -1,4 +1,6 @@
|
|
| 1 |
-
#
|
|
|
|
|
|
|
| 2 |
|
| 3 |
> How likely a dialog response is upvoted 👍 and/or gets replied 💬?
|
| 4 |
|
|
@@ -24,6 +26,13 @@ We considered the following tasks and provided corresponding pretrained models.
|
|
| 24 |
| `human_vs_machine`| ... a machine generated response | [model card](https://huggingface.co/microsoft/DialogRPT-human-vs-machine) |
|
| 25 |
|
| 26 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 27 |
### Contact:
|
| 28 |
Please create an issue on [our repo](https://github.com/golsun/DialogRPT)
|
| 29 |
|
|
|
|
| 1 |
+
# DialogRPT-updown
|
| 2 |
+
|
| 3 |
+
### Dialog Ranking Pretrained Transformers
|
| 4 |
|
| 5 |
> How likely a dialog response is upvoted 👍 and/or gets replied 💬?
|
| 6 |
|
|
|
|
| 26 |
| `human_vs_machine`| ... a machine generated response | [model card](https://huggingface.co/microsoft/DialogRPT-human-vs-machine) |
|
| 27 |
|
| 28 |
|
| 29 |
+
### Examples:
|
| 30 |
+
The `updown` score predicts how likely the response is getting upvoted.
|
| 31 |
+
| Context | Response | `updown` score |
|
| 32 |
+
| :-- | :-- | :--: |
|
| 33 |
+
| I love NLP! | Response: Here’s a free textbook (URL) in case anyone needs it. | 0.613 |
|
| 34 |
+
| I love NLP! | Me too! | 0.111 |
|
| 35 |
+
|
| 36 |
### Contact:
|
| 37 |
Please create an issue on [our repo](https://github.com/golsun/DialogRPT)
|
| 38 |
|