This model is a RoBERTa-based encoder-only student model trained using knowledge distillation for efficient multi-label classification of HPV-related health communication content.

The goal of this model is to enable efficient and scalable information extraction from public health resources while maintaining strong predictive performance.

Task

Multi-label classification of HPV-related content categories.

Model Details

Base model: RoBERTa
Architecture: Encoder-only
Training paradigm: Knowledge Distillation
Objective: Efficient information extraction

Usage

python

from transformers import AutoTokenizer, AutoModelForSequenceClassification

repo = "skhan225/RoBERTa_distilled_HPV"

tokenizer = AutoTokenizer.from_pretrained(repo)
model = AutoModelForSequenceClassification.from_pretrained(repo)

text = "HPV vaccination prevents cervical cancer."
inputs = tokenizer(text, return_tensors="pt")
outputs = model(**inputs)