Distilling Answer-Set Programming Rules from LLMs for Neurosymbolic Visual Question Answering
Theory and Practice of Logic Programming · 31.07.2026 · 1-35
DOI: 10.1017/s1471068426100684 ↗Kurz gefasst
Fachartikel in der Zeitschrift Theory and Practice of Logic Programming, online erschienen am 31.07.2026. Verfasst von NELSON HIGUERA RUIZ und 2 weiteren Autor:innen. Die Arbeit nennt 64 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Visual Question Answering (VQA) is the task of answering questions about images, requiring the integration of multimodal input and reasoning. Modular approaches that incorporate logic-based representations into the reasoning component offer clear advantages over end-to-end trained systems, particularly in terms of interpretability. However, adapting or extending these representations when task requirements change can place a significant burden on developers. To address this challenge, we present an approach for distilling rules from Large Language Models (LLMs). Our method prompts an LLM to extend an initial VQA reasoning theory, expressed as an answer-set program, to meet new requirements of the task. Examples from VQA datasets guide the LLM, validate the results, and help correct erroneous rules by leveraging feedback from the ASP solver. We demonstrate that our approach is effective across diverse VQA datasets. Notably, only a few examples are needed to elicit correct rules from LLMs. Our experiments suggest that rule distillation from LLMs is a promising alternative to traditional data-driven rule learning approaches. Under consideration in Theory and Practice of Logic Programming. 1
Zitieren
APA
RUIZ, N. H., OETSCH, J., & EITER, T. (2026). Distilling Answer-Set Programming Rules from LLMs for Neurosymbolic Visual Question Answering. Theory and Practice of Logic Programming, 1-35. https://doi.org/10.1017/s1471068426100684
Vancouver
RUIZ NH, OETSCH J, EITER T. Distilling Answer-Set Programming Rules from LLMs for Neurosymbolic Visual Question Answering. Theory and Practice of Logic Programming. 2026:1-35. doi:10.1017/s1471068426100684
BibTeX
@article{RUIZ2026,
author = {RUIZ, NELSON HIGUERA and OETSCH, JOHANNES and EITER, THOMAS},
title = {Distilling Answer-Set Programming Rules from LLMs for Neurosymbolic Visual Question Answering},
journal = {Theory and Practice of Logic Programming},
year = {2026},
pages = {1--35},
doi = {10.1017/s1471068426100684}
}
EUVIMED