Emergent Linguistic Structures in Neural Networks are Fragile

Large Language Models (LLMs) have been reported to have strong performance on natural language processing tasks. However, performance metrics such as accuracy do not measure the quality of the model in terms of its ability to robustly represent complex linguistic structure. In this work, we propose a framework and measure of robustness to assess the consistency of linguistic representations against syntax-preserving perturbations. We leverage recent advances in extracting linguistic constructs from LLMs to test the robustness of such structures. Empirically, we study the performance of four LLMs across six different corpora on the proposed robustness measures. We provide evidence that context-free representation (e.g., GloVe) are in some cases competitive with context-dependent representations from modern LLMs (e.g., BERT), yet equally brittle to syntax-preserving manipulations. Emergent syntactic representations in neural networks are brittle, thus our work poses the attention on the risk of comparing such structures to those that are object of a long lasting debate in linguistics.