Spaces:
Runtime error
Runtime error
Shane Weisz
commited on
Commit
•
eca9a50
1
Parent(s):
824a069
Move system description to below
Browse files
app.py
CHANGED
@@ -6,12 +6,12 @@ DECODING_CONFIG = {"max_new_tokens": 100, "min_new_tokens": 20, "no_repeat_ngram
|
|
6 |
|
7 |
TITLE = "Automating Counterspeech in Dialogue Systems"
|
8 |
DESCRIPTION = """
|
9 |
-
|
10 |
-
<br/><br/>
|
11 |
-
**Try it out**: Enter a hate speech comment (or select one of the provided examples below), click Submit, and see if the system generates an appropriate counterspeech response.
|
12 |
"""
|
13 |
|
14 |
ARTICLE = f"""
|
|
|
|
|
15 |
The system is built by fine-tuning [DialoGPT](https://huggingface.co/microsoft/DialoGPT-medium#:~:text=DialoGPT%20is%20a%20SOTA%20large,single%2Dturn%20conversation%20Turing%20test) on the [MultiCONAN](https://github.com/marcoguerini/CONAN#Multitarget-CONAN) dataset, a dataset comprising a set of hate speech inputs and appropriate [counterspeech](https://dangerousspeech.org/counterspeech/) responses produced under the supervision of trained NGO operators from [Stop Hate UK](https://www.stophateuk.org/).
|
16 |
<br/><br/>
|
17 |
**Model:** {DEFAULT_MODEL}<br>
|
|
|
6 |
|
7 |
TITLE = "Automating Counterspeech in Dialogue Systems"
|
8 |
DESCRIPTION = """
|
9 |
+
Enter a hate speech comment (or select one of the provided examples below), click Submit, and see if the system generates an appropriate counterspeech response.
|
|
|
|
|
10 |
"""
|
11 |
|
12 |
ARTICLE = f"""
|
13 |
+
This system has been built by [Shane Weisz](https://shaneweisz.com) for his research project on _Automating Counterspeech in Dialogue Systems_ as part of the [MPhil in Machine Learning and Machine Intelligence](https://www.mlmi.eng.cam.ac.uk/) at Cambridge University. The project is supervised by [Dr Marcus Tomalin](https://www.crassh.cam.ac.uk/about/people/marcus-tomalin/) and forms part of the [Giving Voice to Digital Democracies](https://www.crassh.cam.ac.uk/research/projects-centres/giving-voice-to-digital-democracies/) project on the _The Social Impact of Artificially Intelligent Communications Technology_.
|
14 |
+
<br/><br/>
|
15 |
The system is built by fine-tuning [DialoGPT](https://huggingface.co/microsoft/DialoGPT-medium#:~:text=DialoGPT%20is%20a%20SOTA%20large,single%2Dturn%20conversation%20Turing%20test) on the [MultiCONAN](https://github.com/marcoguerini/CONAN#Multitarget-CONAN) dataset, a dataset comprising a set of hate speech inputs and appropriate [counterspeech](https://dangerousspeech.org/counterspeech/) responses produced under the supervision of trained NGO operators from [Stop Hate UK](https://www.stophateuk.org/).
|
16 |
<br/><br/>
|
17 |
**Model:** {DEFAULT_MODEL}<br>
|