Ideas Have Genomes: Benchmarking Scientific Lineage Reasoning and Lineage-Grounded Idea Generation

2026-07-09

AI Summary

Current AI systems struggle to comprehend and evaluate the genealogical inheritance and evolution of scientific ideas. This work proposes IdeaGene, a framework that models academic papers as typed, evidence-anchored “idea genomes,” and introduces GenomeDiff—a formal mechanism to characterize the processes of idea inheritance, mutation, and recombination. Building upon this framework, the authors construct IG-Bench, a benchmark comprising 1,961 genealogical trajectories, which for the first time enables both genealogical reasoning and genealogy-guided innovative generation. Experiments across 14 large language models reveal that current models achieve only 27.3% accuracy in genealogical reasoning, and structured genealogical context does not consistently improve performance, underscoring both the challenge and necessity of this task.

Abstract

Scientific ideas rarely start from a blank page. They inherit mechanisms, repair known limitations, and recombine pieces of earlier work, much like biological genomes. Current benchmarks still say little about whether AI systems can follow this inheritance structure. We present IdeaGene-Bench (IG-Bench), a benchmark for scientific lineage reasoning and lineage-grounded idea generation. IG-Bench is organized around the IdeaGene framework: each paper or proposal is represented as a set of minimal, typed, evidence-grounded Idea Genome objects, and a GenomeDiff aligns these objects to record inheritance, mutation, loss, external import, and novel insertion under six operational evolutionary dynamics. The benchmark contains 1,961 golden lineage traces, 1,085 curated Idea Genome objects, and 920 pairwise GenomeDiff records across 10 scientific domains. It supports two evaluations. IG-Exam (42 task types, 1,029 instances) tests closed-form lineage reasoning across Idea Genome abstraction, inheritance tracing, evolutionary reasoning, and lineage verification. IG-Arena evaluates generation with a lineage-conditioned Population-Evolution Score(PES), asking whether a proposal can be inserted as a coherent descendant of a given lineage population: it should inherit the right Idea Genome objects, vary meaningfully from nearby work, and offer selection value for future research. Experiments on 14 LLM-based scientists expose a compositional bottleneck. The strongest system reaches only 27.3% exact accuracy on lineage reasoning, and structured lineage context reshuffles system rankings rather than helping every participant uniformly.

Problem

• scientific lineage
• idea generation
• genome analogy
• inheritance structure
• AI reasoning

Innovation

• Idea Genome
• scientific lineage reasoning
• lineage-grounded generation
• GenomeDiff
• evolutionary dynamics