EvoGens: A Population-Based Heuristic Search Framework for Scientific Idea Generation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914616052809728 |
|---|---|
| author | Li, Xu Tu, Hanzhe Li, Xinyi Zhao, Kuncheng Han, Xun Liu, Zhonghui |
| author_facet | Li, Xu Tu, Hanzhe Li, Xinyi Zhao, Kuncheng Han, Xun Liu, Zhonghui |
| contents | Generating novel research ideas is fundamental to scientific progress. While Large Language Models (LLMs) show promise in assisting this process, existing approaches often exhibit semantic convergence, resulting in limited diversity and novelty. To address this, we introduce EvoGens, an evolution-inspired framework that recasts scientific idea generation as an evolutionary search over a population of ideas. EvoGens iteratively applies rank-based mutation with differentiated retrieval planning to incorporate external knowledge, and semantic-aware crossover to fuse complementary concepts for conceptual reorganization. A lightweight evaluation signal guides the selection process, encouraging sustained exploration while mitigating premature convergence. Extensive experiments demonstrate that EvoGens substantially enhances exploration capabilities compared to state-of-the-art baselines. Specifically, it improves the Novelty from 0.1 to 0.4 and the Diversity from 0.24 to 0.55, while maintaining comparable idea quality under the current automatic evaluation protocol. These findings suggest that evolutionary mechanisms can serve as a useful framework for exploration-oriented research ideation, especially for broadening the novelty and diversity of candidate ideas under a shared automatic evaluation setting. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_30961 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | EvoGens: A Population-Based Heuristic Search Framework for Scientific Idea Generation Li, Xu Tu, Hanzhe Li, Xinyi Zhao, Kuncheng Han, Xun Liu, Zhonghui Computation and Language Generating novel research ideas is fundamental to scientific progress. While Large Language Models (LLMs) show promise in assisting this process, existing approaches often exhibit semantic convergence, resulting in limited diversity and novelty. To address this, we introduce EvoGens, an evolution-inspired framework that recasts scientific idea generation as an evolutionary search over a population of ideas. EvoGens iteratively applies rank-based mutation with differentiated retrieval planning to incorporate external knowledge, and semantic-aware crossover to fuse complementary concepts for conceptual reorganization. A lightweight evaluation signal guides the selection process, encouraging sustained exploration while mitigating premature convergence. Extensive experiments demonstrate that EvoGens substantially enhances exploration capabilities compared to state-of-the-art baselines. Specifically, it improves the Novelty from 0.1 to 0.4 and the Diversity from 0.24 to 0.55, while maintaining comparable idea quality under the current automatic evaluation protocol. These findings suggest that evolutionary mechanisms can serve as a useful framework for exploration-oriented research ideation, especially for broadening the novelty and diversity of candidate ideas under a shared automatic evaluation setting. |
| title | EvoGens: A Population-Based Heuristic Search Framework for Scientific Idea Generation |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2605.30961 |