Open access
Date
2023Type
- Journal Article
Abstract
Probing has become a go-to methodology for interpreting and analyzing deep neural models in natural language processing. However, there is still a lack of understanding of the limitations and weaknesses of various types of probes. In this work, we suggest a strategy for input-level intervention on naturalistic sentences. Using our approach, we intervene on the morpho-syntactic features of a sentence, while keeping the rest of the sentence unchanged. Such an intervention allows us to causally probe pre-trained models. We apply our naturalistic causal probing framework to analyze the effects of grammatical gender and number on contextualized representations extracted from three pre-trained models in Spanish, the multilingual versions of BERT, RoBERTa, and GPT-2. Our experiments suggest that naturalistic interventions lead to stable estimates of the causal effects of various linguistic properties. Moreover, our experiments demonstrate the importance of naturalistic causal probing when analyzing pre-trained models. Show more
Permanent link
https://doi.org/10.3929/ethz-b-000614128Publication status
publishedExternal links
Journal / series
Transactions of the Association for Computational LinguisticsVolume
Pages / Article No.
Publisher
MIT PressOrganisational unit
02219 - ETH AI Center / ETH AI Center09682 - Cotterell, Ryan / Cotterell, Ryan
Related publications and datasets
Is new version of: http://hdl.handle.net/20.500.11850/588593
More
Show all metadata