Investigating BERT's Knowledge of Language: Five Analysis Methods with NPIs
This work addresses the challenge of assessing grammatical understanding in NLP models for researchers, though it is incremental as it builds on prior evaluation methods.
The paper tackled the problem of evaluating grammatical knowledge in sentence representation models like BERT, using negative polarity item licensing as a case study, and found that BERT shows significant knowledge but performance varies across five experimental methods.
Though state-of-the-art sentence representation models can perform tasks requiring significant knowledge of grammar, it is an open question how best to evaluate their grammatical knowledge. We explore five experimental methods inspired by prior work evaluating pretrained sentence representation models. We use a single linguistic phenomenon, negative polarity item (NPI) licensing in English, as a case study for our experiments. NPIs like "any" are grammatical only if they appear in a licensing environment like negation ("Sue doesn't have any cats" vs. "Sue has any cats"). This phenomenon is challenging because of the variety of NPI licensing environments that exist. We introduce an artificially generated dataset that manipulates key features of NPI licensing for the experiments. We find that BERT has significant knowledge of these features, but its success varies widely across different experimental methods. We conclude that a variety of methods is necessary to reveal all relevant aspects of a model's grammatical knowledge in a given domain.