Precisely the Point: Adversarial Augmentations for Faithful and Informative Text Generation
This addresses the understudied issue of robustness in text generation for NLP applications, offering a novel method to enhance model reliability, though it is incremental in building on existing adversarial techniques.
The paper tackles the problem of robustness in Seq2Seq text generation models, finding that even SOTA models like BART are vulnerable, leading to degeneration in faithfulness and informativeness, and proposes AdvSeq, an adversarial augmentation framework that significantly improves these aspects across three tasks as shown in experiments.
Though model robustness has been extensively studied in language understanding, the robustness of Seq2Seq generation remains understudied. In this paper, we conduct the first quantitative analysis on the robustness of pre-trained Seq2Seq models. We find that even current SOTA pre-trained Seq2Seq model (BART) is still vulnerable, which leads to significant degeneration in faithfulness and informativeness for text generation tasks. This motivated us to further propose a novel adversarial augmentation framework, namely AdvSeq, for generally improving faithfulness and informativeness of Seq2Seq models via enhancing their robustness. AdvSeq automatically constructs two types of adversarial augmentations during training, including implicit adversarial samples by perturbing word representations and explicit adversarial samples by word swapping, both of which effectively improve Seq2Seq robustness. Extensive experiments on three popular text generation tasks demonstrate that AdvSeq significantly improves both the faithfulness and informativeness of Seq2Seq generation under both automatic and human evaluation settings.