Boosting Prompt-Based Self-Training With Mapping-Free Automatic Verbalizer for Multi-Class Classification
Submission Type: Regular Long Paper
Submission Track: Language Modeling and Analysis of Language Models
Submission Track 2: NLP Applications
Keywords: Prompt-based Learning, Prompt-based fine-tuning, Prompt-based self-training, Verbalizer, Label Word Mapping
TL;DR: Prompt-based self-training for multi-class text classification is under-explored, and a new structure called MAV shows promising self-training performances by automatically extracting word features from MLM predictions.
Abstract: Recently, prompt-based fine-tuning has garnered considerable interest as a core technique for few-shot text classification task. This approach reformulates the fine-tuning objective to align with the Masked Language Modeling (MLM) objective. Leveraging unlabeled data, prompt-based self-training has shown greater effectiveness in binary and three-class classification. However, prompt-based self-training for multi-class classification has not been adequately investigated, despite its significant applicability to real-world scenarios. Moreover, extending current methods to multi-class classification suffers from the verbalizer that extracts the predicted value of manually pre-defined single label word for each class from MLM predictions. Consequently, we introduce a novel, efficient verbalizer structure, named Mapping-free Automatic Verbalizer (MAV). Comprising two fully connected layers, MAV serves as a trainable verbalizer that automatically extracts the requisite word features for classification by capitalizing on all available information from MLM predictions. Experimental results on five multi-class classification datasets indicate MAV's superior self-training efficacy.
Submission Number: 2885
Loading