TY - JOUR
T1 - Abduction of Domain Relationships from Data for VQA
AU - Chowdhury, Al Mehdi Saadat
AU - Shakarian, Paulo
AU - Simari, Gerardo I.
N1 - Publisher Copyright:
© 2025 Open Publishing Association. All rights reserved.
PY - 2025/2/13
Y1 - 2025/2/13
N2 - In this paper, we study the problem of visual question answering (VQA) where the image and query are represented by ASP programs that lack domain data. We provide an approach that is orthogonal and complementary to existing knowledge augmentation techniques where we abduce domain relationships of image constructs from past examples. After framing the abduction problem, we provide a baseline approach, and an implementation that significantly improves the accuracy of query answering yet requires few examples.
AB - In this paper, we study the problem of visual question answering (VQA) where the image and query are represented by ASP programs that lack domain data. We provide an approach that is orthogonal and complementary to existing knowledge augmentation techniques where we abduce domain relationships of image constructs from past examples. After framing the abduction problem, we provide a baseline approach, and an implementation that significantly improves the accuracy of query answering yet requires few examples.
UR - http://www.scopus.com/inward/record.url?scp=85218633706&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85218633706&partnerID=8YFLogxK
U2 - 10.4204/EPTCS.416.15
DO - 10.4204/EPTCS.416.15
M3 - Conference article
AN - SCOPUS:85218633706
SN - 2075-2180
VL - 416
SP - 168
EP - 174
JO - Electronic Proceedings in Theoretical Computer Science, EPTCS
JF - Electronic Proceedings in Theoretical Computer Science, EPTCS
T2 - 40th International Conference on Logic Programming, ICLP 2024
Y2 - 14 October 2024 through 17 October 2024
ER -