CHIMERA: A Knowledge Base of Idea Recombination in Scientific Literature

ACL ARR 2025 July Submission90 Authors

Abstract:

A hallmark of human innovation is *recombination*---the creation of novel ideas by integrating elements from existing concepts and mechanisms. In this work, we introduce CHIMERA, a large-scale Knowledge Base (KB) of over 28K recombination examples automatically mined from the scientific literature. CHIMERA enables large-scale empirical analysis of how scientists recombine concepts and draw inspiration from different areas, and enables training models that propose novel, cross-disciplinary research directions.
To construct this KB, we define a new information extraction task: identifying recombination instances in scientific abstracts. We curate a high-quality, expert-annotated dataset and use it to fine-tune a large language model, which we apply to a broad corpus of AI papers.
We showcase the utility of CHIMERA through two applications. First, we analyze patterns of recombination across AI subfields. Second, we train a scientific hypothesis generation model using the KB, showing that it can propose novel research directions that researchers rate as inspiring. We release our data and code at https://anonymous.4open.science/r/CHIMERA-0510.

Paper Type: Long

Research Area: NLP Applications

Research Area Keywords: knowledge graphs

Contribution Types: NLP engineering experiment, Publicly available software and/or pre-trained models, Data resources, Data analysis

Languages Studied: English

Previous URL: https://openreview.net/forum?id=EKHv05vmeG

Reassignment Request Area Chair: Yes, I want a different area chair for our submission

Reassignment Request Reviewers: Yes, I want a different set of reviewers

Justification For Not Keeping Action Editor Or Reviewers: We request reassignment of both reviewers and the meta-reviewer. Reviews had the following issues: shallow engagement with the paper, lack of demonstrated expertise (including factual misunderstandings of standard methodology and dismissing well-established practices), signs of AI-generated content, and a failure to acknowledge key clarifications and evidence provided in our rebuttal, including the qualifications of our annotators and the clearly stated motivation, novelty, and value of our dataset.

A1 Limitations Section: This paper has a limitations section.

A2 Potential Risks: N/A

B Use Or Create Scientific Artifacts: Yes

B1 Cite Creators Of Artifacts: Yes

B1 Elaboration: Section 3, Appendix B.2, Appendix E.2, Section 5

B2 Discuss The License For Artifacts: N/A

B3 Artifact Use Consistent With Intended Use: Yes

B3 Elaboration: Ethical Considerations

B4 Data Contains Personally Identifying Info Or Offensive Content: Yes

B4 Elaboration: We collect scientific data (the language is professional). We do not disclose annotators personal information, as discussed in our Ethics Statement.

B5 Documentation Of Artifacts: Yes

B5 Elaboration: Section 3, tables 2, 3 + text describing the knowledge base.

B6 Statistics For Data: Yes

B6 Elaboration: Section 3, Table 2, 3 and Section 5, Table 5

C Computational Experiments: Yes

C1 Model Size And Budget: Yes

C1 Elaboration: Appendix B.2 (extraction baselines implementation), Appendix E.2 (prediction baselines implementation)

C2 Experimental Setup And Hyperparameters: Yes

C2 Elaboration: Section 4.1 (experimental settings). Additional implementation details, such as hyper parameters, are in Appendix B.2, Appendix E.2. We also provide a highly documented code repository for reproducing our experiments - https://anonymous.4open.science/r/CHIMERA-0510

C3 Descriptive Statistics: Yes

C3 Elaboration: Section 4 (results), Section 4.2 (Extraction Results)

C4 Parameters For Packages: Yes

C4 Elaboration: We provide implementation details including specific parameters in Appendix B.2 and Appendix E.2. We also provide a highly documented code repository for reproducing our experiments - https://anonymous.4open.science/r/CHIMERA-0510

D Human Subjects Including Annotators: Yes

D1 Instructions Given To Participants: Yes

D1 Elaboration: Section 3.1 (recombination mining) references our annotations guidelines for the extraction task. We provide our user study guidelines in Appendix F.

D2 Recruitment And Payment: Yes

D2 Elaboration: We detail our recruitment process in Section 3.1. Annotators' payment is discussed in the Ethical Considerations section

D3 Data Consent: Yes

D3 Elaboration: Our Section 3.1 (recombination mining) references our annotations guidelines for the extraction task. We provide our user study guidelines in Appendix F. Both describe the intended use of the collected data.

D4 Ethics Review Board Approval: N/A

D5 Characteristics Of Annotators: Yes

D5 Elaboration: The annotators characteristics are described in Section 3.1. Section 5.1 describes the characteristics of the user-study participants.

E Ai Assistants In Research Or Writing: Yes

E1 Information About Use Of Ai Assistants: Yes

E1 Elaboration: Ethical Considerations

Author Submission Checklist: yes

Submission Number: 90