CHIMERA: A Knowledge Base of Idea Recombination in Scientific Literature

2025-05-27

AI Summary

This study empirically investigates how scientists recombine scientific concepts across domains to stimulate innovation and support supervised cross-domain hypothesis generation. Method: We introduce “scientific concept recombination” as a novel cognitive modeling task and construct CHIMERA—the first large-scale, publicly available knowledge base—by automatically extracting and manually refining over 28,000 real-world recombination instances from AI-domain paper abstracts. We design an LLM-augmented information extraction framework for high-precision identification of concept pairs and their source domains, conduct statistical analyses to characterize cross-domain recombination patterns, and train supervised models to generate verifiable scientific hypotheses. Contribution/Results: Experiments demonstrate that hypotheses generated by our model receive strong validation and endorsement from domain experts. All code, data, and the CHIMERA knowledge base are fully open-sourced to advance AI-augmented scientific discovery research.

Technology Category

• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning
• Application Domains: Humanities & Computational Social Science
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

A hallmark of human innovation is the process of recombination -- creating original ideas by integrating elements of existing mechanisms and concepts. In this work, we automatically mine the scientific literature and build CHIMERA: a large-scale knowledge base (KB) of recombination examples. CHIMERA can be used to empirically explore at scale how scientists recombine concepts and take inspiration from different areas, or to train supervised machine learning models that learn to predict new creative cross-domain directions. To build this KB, we present a novel information extraction task of extracting recombination from scientific paper abstracts, collect a high-quality corpus of hundreds of manually annotated abstracts, and use it to train an LLM-based extraction model. The model is applied to a large corpus of papers in the AI domain, yielding a KB of over 28K recombination examples. We analyze CHIMERA to explore the properties of recombination in different subareas of AI. Finally, we train a scientific hypothesis generation model using the KB, which predicts new recombination directions that real-world researchers find inspiring. Our data and code are available at https://github.cs.huji.ac.il/tomhope-lab/CHIMERA

Problem

• Mining scientific literature for idea recombination examples
• Building a knowledge base to explore cross-domain inspiration
• Training models to predict creative scientific directions

Innovation

• LLM-based extraction model for recombination examples
• Large-scale knowledge base with 28K examples
• Scientific hypothesis generation model training