Robustness of Demonstration-based Learning Under Limited Data Scenario
This work addresses the robustness of demonstration-based learning for few-shot NER, providing insights into its mechanisms, but it is incremental as it builds on existing methods without introducing new paradigms.
The paper investigates why demonstration-based learning improves few-shot named entity recognition (NER) under limited data, finding that even random token demonstrations enhance model performance, with length and token relevance being key factors, and demonstrations increase model confidence in superficial patterns.
Demonstration-based learning has shown great potential in stimulating pretrained language models' ability under limited data scenario. Simply augmenting the input with some demonstrations can significantly improve performance on few-shot NER. However, why such demonstrations are beneficial for the learning process remains unclear since there is no explicit alignment between the demonstrations and the predictions. In this paper, we design pathological demonstrations by gradually removing intuitively useful information from the standard ones to take a deep dive of the robustness of demonstration-based sequence labeling and show that (1) demonstrations composed of random tokens still make the model a better few-shot learner; (2) the length of random demonstrations and the relevance of random tokens are the main factors affecting the performance; (3) demonstrations increase the confidence of model predictions on captured superficial patterns. We have publicly released our code at https://github.com/SALT-NLP/RobustDemo.