The Ramon Llull's Thinking Machine for Automated Ideation

TL;DR

This paper develops a modern AI-based ideation tool inspired by Ramon Llull's medieval combinatorial framework, enabling diverse and relevant research idea generation through symbolic recombination of high-level scientific elements.

Contribution

It introduces a novel LLM-driven thinking machine that uses thematic, domain, and methodological axes for automated research idea generation, grounded in literature.

Findings

Generated diverse, relevant research ideas from curated combinations.

Demonstrated the tool's ability to augment scientific creativity.

Showed the approach's interpretability and potential for collaborative ideation.

Abstract

This paper revisits Ramon Llull's Ars combinatoria - a medieval framework for generating knowledge through symbolic recombination - as a conceptual foundation for building a modern Llull's thinking machine for research ideation. Our approach defines three compositional axes: Theme (e.g., efficiency, adaptivity), Domain (e.g., question answering, machine translation), and Method (e.g., adversarial training, linear attention). These elements represent high-level abstractions common in scientific work - motivations, problem settings, and technical approaches - and serve as building blocks for LLM-driven exploration. We mine elements from human experts or conference papers and show that prompting LLMs with curated combinations produces research ideas that are diverse, relevant, and grounded in current literature. This modern thinking machine offers a lightweight, interpretable tool for augmenting scientific creativity and suggests a path toward collaborative ideation between humans and AI.