Solving Abstract Reasoning Problems With Neurosymbolic Program Synthesis and Task Generation
Neurosymbolic Artificial Intelligence · 2026-01 · 2
DOI: 10.1177/29498732261419309 ↗Kurz gefasst
Fachartikel in der Zeitschrift Neurosymbolic Artificial Intelligence, online erschienen am 02.03.2026. Verfasst von Jakub Bednarek und 1 weiteren Autor:innen. Die Arbeit nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
The ability to think abstractly and reason by analogy is a prerequisite to rapidly adapt to new conditions, tackle newly encountered problems by decomposing them, and synthesize knowledge to solve problems comprehensively. We present TransCoder, a method for solving abstract problems based on neural program synthesis, and conduct a comprehensive analysis of decisions made by the generative module of the proposed architecture. At the core of TransCoder is a typed domain-specific language, designed to facilitate feature engineering and abstract reasoning. In training, we use the programs that failed to solve tasks to generate new tasks and gather them in a synthetic dataset. As each synthetic task created in this way has a known associated program (solution), the model is trained on them in supervised mode. Solutions are represented in a transparent programmatic form, which can be inspected and verified. We demonstrate TransCoder’s performance using the Abstract Reasoning Corpus dataset, for which our framework generates tens of thousands of synthetic problems with corresponding solutions and facilitates systematic progress in learning.
Zitieren
APA
Bednarek, J., & Krawiec, K. (2026). Solving Abstract Reasoning Problems With Neurosymbolic Program Synthesis and Task Generation. Neurosymbolic Artificial Intelligence, 2. https://doi.org/10.1177/29498732261419309
Vancouver
Bednarek J, Krawiec K. Solving Abstract Reasoning Problems With Neurosymbolic Program Synthesis and Task Generation. Neurosymbolic Artificial Intelligence. 2026;2. doi:10.1177/29498732261419309
BibTeX
@article{Bednarek2026,
author = {Bednarek, Jakub and Krawiec, Krzysztof},
title = {Solving Abstract Reasoning Problems With Neurosymbolic Program Synthesis and Task Generation},
journal = {Neurosymbolic Artificial Intelligence},
year = {2026},
volume = {2},
doi = {10.1177/29498732261419309}
}
EUVIMED