visolex / phobert-hsd

PhoBERT‑HSD: Hate Speech Detection for Vietnamese Text

Fine‑tuned from vinai/phobert-base on the VN‑HSD dataset.

Model Details

• Base Model : vinai/phobert-base
• Dataset : VN‑HSD (ViSoLex‑HSD unified hate speech corpus)
• Fine‑tuning : HuggingFace Transformers

Hyperparameters

• Batch size: 32
• Learning rate: 5e-5
• Epochs: 100
• Max sequence length: 256

Results

• Accuracy : <INSERT_ACCURACY>
• F1 Score : <INSERT_F1_SCORE>

Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification

tokenizer = AutoTokenizer.from_pretrained("visolex/phobert-hsd")
model = AutoModelForSequenceClassification.from_pretrained("visolex/phobert-hsd")

text = "Đừng nói những lời thô tục như vậy!"
inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=256)
pred = model(**inputs).logits.argmax(dim=-1).item()
print(f"Label: {['CLEAN','OFFENSIVE','HATE'][pred]}")