File size: 716 Bytes
12a5bfe
 
a1a7267
 
12a5bfe
a1a7267
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
---
license: cc-by-4.0
language:
- en
---


# Counter-Speech Classifier

<!-- Provide a quick summary of what the model is/does. -->

The BERT-based counter-speech classifier is finetuned on the [CONAN](https://github.com/marcoguerini/CONAN) dataset for classifying whether a response is counter-speech, based on the counter-argument classifier [ThinkCERCA/counterargument_hugging](https://huggingface.co/ThinkCERCA/counterargument_hugging)
## Uses

<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
The model is intended for classifying LM-generated dialogue responses, evaluating their validity as counter-speech.