This model encodes texts and candidate labels into a shared embedding space using BERT, enabling classification into arbitrary categories without retraining for new labels.

Training Details

Parameter	Value
Base model	`bert-base-uncased`
Model variant	`late_interaction`
Training steps	1000
Batch size	2
Learning rate	2e-05
Trainable params	109,580,672
Training time	354.7s

Dataset

Trained on polodealvarado/zeroshot-classification.

Evaluation Results

Metric	Score
Precision	0.8546
Recall	0.9686
F1 Score	0.9081

Usage

python

from models.late_interaction import LateInteractionModel

model = LateInteractionModel.from_pretrained("polodealvarado/late_interaction")

predictions = model.predict(
    texts=["The stock market crashed yesterday."],
    labels=[["Finance", "Sports", "Biology", "Economy"]],
)
print(predictions)
# [{"text": "...", "scores": {"Finance": 0.98, "Economy": 0.85, ...}}]