Reconstruction Probing

TL;DR

Reconstruction probing is a new method for analyzing contextualized representations in masked language models by comparing reconstruction probabilities with and without contextual information, revealing how context influences token reconstruction.

Contribution

This paper introduces reconstruction probing, a novel analysis technique for understanding the impact of contextualization in masked language models, including its effects on token reconstructability.

Findings

Contextualization enhances reconstructability of nearby tokens.

Static and positional embeddings contribute significantly to contextualization effects.

Reconstruction probabilities quantify the influence of context on token representations.

Abstract

We propose reconstruction probing, a new analysis method for contextualized representations based on reconstruction probabilities in masked language models (MLMs). This method relies on comparing the reconstruction probabilities of tokens in a given sequence when conditioned on the representation of a single token that has been fully contextualized and when conditioned on only the decontextualized lexical prior of the model. This comparison can be understood as quantifying the contribution of contextualization towards reconstruction -- the difference in the reconstruction probabilities can only be attributed to the representational change of the single token induced by contextualization. We apply this analysis to three MLMs and find that contextualization boosts reconstructability of tokens that are close to the token being reconstructed in terms of linear and syntactic distance. Furthermore, we extend our analysis to finer-grained decomposition of contextualized representations, and we find that these boosts are largely attributable to static and positional embeddings at the input layer.