Decoding Probing: Revealing Internal Linguistic Structures in Neural Language Models using Minimal Pairs

arXiv:2403.17299v1 Announce Type: new
Abstract: Inspired by cognitive neuroscience studies, we introduce a novel `decoding probing' method that uses minimal pairs benchmark (BLiMP) to probe internal linguistic characteristics in neural language models layer by layer. By treating the language model as the `brain' and its representations as `neural activations', we decode grammaticality labels of minimal pairs from the intermediate layers' representations. This approach reveals: 1) Self-supervised language models capture abstract linguistic structures in intermediate layers that GloVe and RNN language …

abstract arxiv benchmark brain cognitive cs.cl decoding language language model language models layer neuroscience novel probe q-bio.nc studies type