The current and future role of visual question answering in eXplainable artificial intelligence.

Caro-Martinez, Marta; Wijekoon, Anjana; Diaz-Agudo, Belén; Recio-Garcia, Juan A.

Abstract

Over the last few years, we have seen how the interest of the computer science research community on eXplainable Artificial Intelligence has grown in leaps and bounds. The reason behind this rise is the use of Artificial Intelligence in many daily life tasks, and the consequent necessity of people to understand the intelligent systems' behaviour. Computer vision-related tasks are not an exception, for example, Visual Question Answering tasks. The Artificial Intelligence models that carry out this specific task make an effort to answer questions about what we can watch in a particular image. In this work, we review the existing work about eXplainable Artificial Intelligence on Visual Question Answering which is a problem on which there is still much work to be done. Moreover, we open the discussion about the challenges to overcome regarding this topic, like the future role of Visual Question Answering to address eXplainable Artificial Intelligence issues or difficulties.

Citation

CARO-MARTINEZ, M., WIJEKOON, A., DIAZ-AGUDO, B. and RECIO-GARCIA, J.A. 2023. The current and future role of visual question answering in eXplainable artificial intelligence. In Malburg, L. and Verma, D. (eds.) Workshop proceedings of the 31st International conference on case-based reasoning (ICCBR-WS 2023), 17 July 2023, Aberdeen, UK. CEUR workshop proceedings, 3438. Aachen: CEUR-WS [online], pages 172-183. Available from: https://ceur-ws.org/Vol-3438/paper_13.pdf

• Presentation Conference Type: Conference Paper (published)
• Conference Name: Workshops of the 31st International conference on case-based reasoning (ICCBR-WS 2023)
• Start Date: Jul 17, 2023
• Acceptance Date: Jun 16, 2023
• Online Publication Date: Jul 17, 2023
• Publication Date: Aug 7, 2023
• Deposit Date: Aug 21, 2023
• Publicly Available Date: Aug 21, 2023
• Publisher: CEUR-WS
• Peer Reviewed: Peer Reviewed
• Pages: 172-183
• Series Title: CEUR workshop proceedings
• Series Number: 3438
• Series ISSN: 1613-0073
• Keywords: eXplainable artificial intelligence; XAI; Visual question answering; VQA; Natural language processing; NLP; Computer vision
• Public URL: https://rgu-repository.worktribe.com/output/2048580
• Publisher URL: https://ceur-ws.org/Vol-3438/paper_13.pdf