Anchoring Fine-tuning of Sentence Transformer with Semantic Label Information for Efficient Truly Few-shot Classification
Amalie Brogaard Pauli, Leon Derczynski, Ira Assent
Abstract
Few-shot classification is a powerful technique, but training requires substantial computing power and data. We propose an efficient method with small model sizes and less training data with only 2-8 training instances per class. Our proposed method, AncSetFit, targets low data scenarios by anchoring the task and label information through sentence embeddings in fine-tuning a Sentence Transformer model. It uses contrastive learning and a triplet loss to enforce training instances of a class to be closest to its own textual semantic label information in the embedding space - and thereby learning to embed different class instances more distinct. AncSetFit obtains strong performance in data-sparse scenarios compared to existing methods across SST-5, Emotion detection, and AG News data, even with just two examples per class.
BibTeX
@inproceedings{
pauli2023anchoring,
title={Anchoring Fine-tuning of Sentence Transformer with Semantic Label Information for Efficient Truly Few-shot Classification},
author={Amalie Brogaard Pauli and Leon Derczynski and Ira Assent},
booktitle={The 2023 Conference on Empirical Methods in Natural Language Processing},
year={2023},
url={https://openreview.net/forum?id=JHd4FSJSC5}
}