The King is Naked: on the Notion of Robustness for Natural Language Processing
This addresses the problem of defining and measuring robustness in NLP for researchers and practitioners, offering a more linguistically grounded approach.
The paper argues that the classical notion of adversarial robustness is inadequate for NLP because it overlooks key linguistic phenomena, and proposes semantic robustness as a better alternative, showing it can improve performance on complex tasks where classical methods fail.
There is growing evidence that the classical notion of adversarial robustness originally introduced for images has been adopted as a de facto standard by a large part of the NLP research community. We show that this notion is problematic in the context of NLP as it considers a narrow spectrum of linguistic phenomena. In this paper, we argue for semantic robustness, which is better aligned with the human concept of linguistic fidelity. We characterize semantic robustness in terms of biases that it is expected to induce in a model. We study semantic robustness of a range of vanilla and robustly trained architectures using a template-based generative test bed. We complement the analysis with empirical evidence that, despite being harder to implement, semantic robustness can improve performance %gives guarantees for on complex linguistic phenomena where models robust in the classical sense fail.