Instructions to use Hello-SimpleAI/chatgpt-detector-roberta with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Hello-SimpleAI/chatgpt-detector-roberta with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="Hello-SimpleAI/chatgpt-detector-roberta")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("Hello-SimpleAI/chatgpt-detector-roberta") model = AutoModelForSequenceClassification.from_pretrained("Hello-SimpleAI/chatgpt-detector-roberta", device_map="auto") - Inference
- Notebooks
- Google Colab
- Kaggle
Production Implementation: Veridraft AI Detector Pro (Hybrid Context + Burstiness Calibration)
#5
by truecopythereal - opened
Hey everyone! Just wanted to share a production implementation built on top of this model: Veridraft AI Detector Pro.
To reduce false positives on short isolated sentences and address line-break issues in uploaded documents, I built a hybrid pipeline wrapping Hello-SimpleAI/chatgpt-detector-roberta:
- Global Context Scan: Evaluates document-level probability alongside sentence windowing.
- Burstiness Calibration: Weights risk against standard deviation in sentence length (CV = std_dev / mean).
- Normalized Parsing: Cleans PDF and Docx text extractions to match raw text inputs.
π Live App: https://thetruecopy-veridraft-detector-app-thguvb.streamlit.app
Would love to hear any thoughts from developers working with this checkpoint or suggestions on model tuning!