Probing Classifiers: Promises, Shortcomings, and Alternatives

45 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Conditional probing: measuring usable information beyond a baseline (2021, background): “Probing analyzes a representation by us-ing it as input into a supervised classiﬁer, which is trained to predict a property, such as part-of-speech (Shi et al., 2016; Ettinger et al., 2016; Alain and Bengio, 2016; Adi et al., 2017; Belinkov, 2021).”
Post-hoc Interpretability for Neural NLP: A Survey (2021, background): “If this concern is valid, it would mean that the probing methods do not provide faithful explanations (Belinkov, 2021).”
Masked Language Modeling and the Distributional Hypothesis: Order Word Matters Pre-training for Little (2021, background): “(Belinkov, 2021) including how complex probes should be.”
Can Language Models Encode Perceptual Structure Without Grounding? A Case Study in Color (2021, background): “the various methodological weaknesses associated with the probing paradigm (Belinkov, 2021).”
The Hitchhiker’s Guide to Computational Linguistics in Suicide Prevention (2021, methodology): “Following are three representative interpretational techniques that can be used to achieve these tasks (for additional inter-pretational methods, see Belinkov, 2021).”