Explore Spurious Correlations at the Concept Level in Language Models for Text Classification

arXiv:2311.08648v3 Announce Type: replace
Abstract: Language models (LMs) have achieved notable success in numerous NLP tasks, employing both fine-tuning and in-context learning (ICL) methods. While language models demonstrate exceptional performance, they face robustness challenges due to spurious correlations arising from imbalanced label distributions in training data or ICL exemplars. Previous research has primarily concentrated on word, phrase, and syntax features, neglecting the concept level, often due to the absence of concept labels and difficulty in identifying conceptual content in input …

abstract arxiv challenges classification concept context correlations cs.ai cs.cl data explore face fine-tuning in-context learning language language models lms nlp performance robustness success tasks text text classification training training data type