FlowPIE: Test-Time Scientific Idea Evolution with Flow-Guided Literature Exploration

TL;DR

FlowPIE introduces a dynamic, flow-guided approach to scientific idea generation that co-evolves literature exploration and idea development, enhancing diversity and quality over static methods.

Contribution

It presents a novel framework combining flow-guided Monte Carlo Tree Search and idea evolution techniques for improved scientific idea generation.

Findings

FlowPIE achieves higher novelty, feasibility, and diversity in generated ideas.

The framework effectively mitigates information cocoons from static literature reliance.

FlowPIE enables reward scaling during test time for better idea quality.

Abstract

Scientific idea generation (SIG) is critical to AI-driven autonomous research, yet existing approaches are often constrained by a static retrieval-then-generation paradigm, leading to homogeneous and insufficiently divergent ideas. In this work, we propose FlowPIE, a tightly coupled retrieval-generation framework that treats literature exploration and idea generation as a co-evolving process. FlowPIE expands literature trajectories via a flow-guided Monte Carlo Tree Search (MCTS) inspired by GFlowNets, using the quality of current ideas assessed by an LLM-based generative reward model (GRM) as a supervised signal to guide adaptive retrieval and construct a diverse, high-quality initial population. Based on this population, FlowPIE models idea generation as a test-time idea evolution process, applying selection, crossover, and mutation with the isolation island paradigm and GRM-based fitness computation to incorporate cross-domain knowledge. It effectively mitigates the information cocoons arising from over-reliance on parametric knowledge and static literature. Extensive evaluations demonstrate that FlowPIE consistently produces ideas with higher novelty, feasibility and diversity compared to strong LLM-based and agent-based frameworks, while enabling reward scaling during test time.