TU Darmstadt / ULB / TUbiblio

UKP-SQuARE v2: Explainability and Adversarial Attacks for Trustworthy QA

Sachdeva, Rachneet ; Puerto San Roman, Haritz ; Baumgärtner, Tim ; Tariverdian, Sewin ; Zhang, Hao ; Wang, Kexin ; Saadi, Hossain Shaikh ; Ribeiro, Leonardo F. R. ; Gurevych, Iryna (2022)
UKP-SQuARE v2: Explainability and Adversarial Attacks for Trustworthy QA.
2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing: System Demonstrations. Taipei, Taiwan (20.-23.11.2022)
Conference or Workshop Item, Bibliographie

Abstract

Question Answering (QA) systems are increasingly deployed in applications where they support real-world decisions. However, state-of-the-art models rely on deep neural networks, which are difficult to interpret by humans. Inherently interpretable models or post hoc explainability methods can help users to comprehend how a model arrives at its prediction and, if successful, increase their trust in the system. Furthermore, researchers can leverage these insights to develop new methods that are more accurate and less biased. In this paper, we introduce SQuARE v2, the new version of SQuARE, to provide an explainability infrastructure for comparing models based on methods such as saliency maps and graph-based explanations. While saliency maps are useful to inspect the importance of each input token for the model’s prediction, graph-based explanations from external Knowledge Graphs enable the users to verify the reasoning behind the model prediction. In addition, we provide multiple adversarial attacks to compare the robustness of QA models. With these explainability methods and adversarial attacks, we aim to ease the research on trustworthy QA models. SQuARE is available on https://square.ukp-lab.de.

Item Type: Conference or Workshop Item
Erschienen: 2022
Creators: Sachdeva, Rachneet ; Puerto San Roman, Haritz ; Baumgärtner, Tim ; Tariverdian, Sewin ; Zhang, Hao ; Wang, Kexin ; Saadi, Hossain Shaikh ; Ribeiro, Leonardo F. R. ; Gurevych, Iryna
Type of entry: Bibliographie
Title: UKP-SQuARE v2: Explainability and Adversarial Attacks for Trustworthy QA
Language: English
Date: 30 November 2022
Publisher: ACL
Book Title: Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing: System Demonstrations
Event Title: 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing: System Demonstrations
Event Location: Taipei, Taiwan
Event Dates: 20.-23.11.2022
URL / URN: https://aclanthology.org/2022.aacl-demo.4
Abstract:

Question Answering (QA) systems are increasingly deployed in applications where they support real-world decisions. However, state-of-the-art models rely on deep neural networks, which are difficult to interpret by humans. Inherently interpretable models or post hoc explainability methods can help users to comprehend how a model arrives at its prediction and, if successful, increase their trust in the system. Furthermore, researchers can leverage these insights to develop new methods that are more accurate and less biased. In this paper, we introduce SQuARE v2, the new version of SQuARE, to provide an explainability infrastructure for comparing models based on methods such as saliency maps and graph-based explanations. While saliency maps are useful to inspect the importance of each input token for the model’s prediction, graph-based explanations from external Knowledge Graphs enable the users to verify the reasoning behind the model prediction. In addition, we provide multiple adversarial attacks to compare the robustness of QA models. With these explainability methods and adversarial attacks, we aim to ease the research on trustworthy QA models. SQuARE is available on https://square.ukp-lab.de.

Uncontrolled Keywords: UKP_p_square, UKP_p_qa_sci_inf
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Ubiquitous Knowledge Processing
Date Deposited: 30 Nov 2022 13:49
Last Modified: 06 Jul 2023 09:25
PPN: 503807400
Export:
Suche nach Titel in: TUfind oder in Google
Send an inquiry Send an inquiry

Options (only for editors)
Show editorial Details Show editorial Details