Generating Adversarial Examples for Topic-dependent Argument Classification - Archive ouverte HAL Access content directly
Conference Papers Year : 2020

Generating Adversarial Examples for Topic-dependent Argument Classification


In the last years, several empirical approaches have been proposed to tackle argument mining tasks, e.g., argument classification, relation prediction, argument synthesis. These approaches rely more and more on language models (e.g., BERT) to boost their performance. However, these language models require a lot of training data, and size is often a drawback of the available argument mining data sets. The goal of this paper is to assess the robustness of these language models for the argument classification task. More precisely, the aim of the current work is twofold: first, we generate adversarial examples addressing linguistic perturbations in the original sentences, and second, we improve the robustness of argument classification models using adversarial training. Two empirical evaluations are addressed relying on standard datasets for AM tasks, whilst the generated adversarial examples are qualitatively evaluated through a user study. Results prove the robust-ness of BERT for the argument classification task, yet highlighting that it is not invulnerable to simple linguistic perturbations in the input data.
Fichier principal
Vignette du fichier
63_Mayer.pdf (139.01 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-02933266 , version 1 (08-09-2020)


  • HAL Id : hal-02933266 , version 1


Tobias Mayer, Santiago Marro, Elena Cabrio, Serena Villata. Generating Adversarial Examples for Topic-dependent Argument Classification. COMMA 2020 - 8th International Conference on Computational Models of Argument, Sep 2020, Perugia, Italy. ⟨hal-02933266⟩
245 View
210 Download


Gmail Facebook X LinkedIn More