PETapter: Leveraging PET-style classification heads for modular few-shot parameter-efficient fine-tuning
This work addresses the challenge for researchers in specialized scientific domains who lack resources to fine-tune large models, offering an incremental improvement in few-shot learning efficiency.
The paper tackles the problem of data scarcity and high computational costs in fine-tuning large language models for specialized domains by proposing PETapter, a method that combines parameter-efficient fine-tuning with PET-style classification heads. The result is comparable performance to full fine-tuning on three NLP benchmarks and a real-world dataset, with greater reliability, higher parameter efficiency, and improved modularity.
Few-shot learning and parameter-efficient fine-tuning (PEFT) are crucial to overcome the challenges of data scarcity and ever growing language model sizes. This applies in particular to specialized scientific domains, where researchers might lack expertise and resources to fine-tune high-performing language models to nuanced tasks. We propose PETapter, a novel method that effectively combines PEFT methods with PET-style classification heads to boost few-shot learning capabilities without the significant computational overhead typically associated with full model training. We validate our approach on three established NLP benchmark datasets and one real-world dataset from communication research. We show that PETapter not only achieves comparable performance to full few-shot fine-tuning using pattern-exploiting training (PET), but also provides greater reliability and higher parameter efficiency while enabling higher modularity and easy sharing of the trained modules, which enables more researchers to utilize high-performing NLP-methods in their research.