The ERINIA team is organising a shared task on Robustness of Credibility Assessment with Adversarial Examples (InCrediblAE), which is taking place as Task 6 at the CheckThat! 2024 Evaluation Lab, whose the results will be presented in a workshop at CLEF 2024 conference in Grenoble, France.

The participants’ aim will is to create ‘adversarial examples’ by making small modifications to each text snippet in the attack dataset that change the victim classifier’s decision without altering the text meaning. The evaluation is based on the BODEGA framework, designed within our project. See the task website for all the details.