Can Prompt Probe Pretrained Language Models? Understanding the Invisible Risks from a Causal View
This addresses risks in evaluating and applying pretrained language models, which is crucial for NLP researchers and practitioners, though it is incremental as it builds on existing probing methods.
The paper tackles the problem of inaccurate and unreliable prompt-based probing for evaluating pretrained language models by investigating it from a causal view, identifying three critical biases and proposing debiasing via causal intervention to improve reliability.
Prompt-based probing has been widely used in evaluating the abilities of pretrained language models (PLMs). Unfortunately, recent studies have discovered such an evaluation may be inaccurate, inconsistent and unreliable. Furthermore, the lack of understanding its inner workings, combined with its wide applicability, has the potential to lead to unforeseen risks for evaluating and applying PLMs in real-world applications. To discover, understand and quantify the risks, this paper investigates the prompt-based probing from a causal view, highlights three critical biases which could induce biased results and conclusions, and proposes to conduct debiasing via causal intervention. This paper provides valuable insights for the design of unbiased datasets, better probing frameworks and more reliable evaluations of pretrained language models. Furthermore, our conclusions also echo that we need to rethink the criteria for identifying better pretrained language models. We openly released the source code and data at https://github.com/c-box/causalEval.