Deep Fake Text Detection
Collection
1 item
•
Updated
Finetuned RoBERTa-base model for detecting AI generated English texts.
See FakespotAILabs/ApolloDFT for more details and a technical report of the model and experiments we conducted.
You can use this model directly with a pipeline.
For better performance, you should apply the clean_text
function in utils.py.
from transformers import pipeline
from utils import clean_text
classifier = pipeline(
"text-classification",
model="fakespot-ai/roberta-base-ai-text-detection-v1"
)
# single text
text = "text 1"
classifier(clean_text(text))
[
{
'label': str,
'score': float
}
]
# list of texts
texts = ["text 1", "text 2"]
classifier([clean_text(t) for t in texts])
[
{
'label': str,
'score': float
},
{
'label': str,
'score': float
}
]
Base model
FacebookAI/roberta-base