Abductive Symbolic Solver on Abstraction and Reasoning Corpus
This work addresses the problem of enhancing AI's logical and human-like reasoning capabilities in visual tasks, representing an incremental improvement over previous grid-focused methods.
The paper tackles the challenge of improving AI reasoning on the Abstraction and Reasoning Corpus (ARC) by proposing a framework that uses abductive reasoning to symbolically represent data into a knowledge graph, extract core knowledge, and narrow the solution space for logical solutions.
This paper addresses the challenge of enhancing artificial intelligence reasoning capabilities, focusing on logicality within the Abstraction and Reasoning Corpus (ARC). Humans solve such visual reasoning tasks based on their observations and hypotheses, and they can explain their solutions with a proper reason. However, many previous approaches focused only on the grid transition and it is not enough for AI to provide reasonable and human-like solutions. By considering the human process of solving visual reasoning tasks, we have concluded that the thinking process is likely the abductive reasoning process. Thus, we propose a novel framework that symbolically represents the observed data into a knowledge graph and extracts core knowledge that can be used for solution generation. This information limits the solution search space and helps provide a reasonable mid-process. Our approach holds promise for improving AI performance on ARC tasks by effectively narrowing the solution space and providing logical solutions grounded in core knowledge extraction.