Papers
arxiv:2603.29557

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

Published on Mar 31
· Submitted by
Mathsion Wong
on Apr 1
Authors:
,
,
,
,
,
,
,

Abstract

FlowPIE presents a novel retrieval-generation framework for scientific idea generation that uses flow-guided Monte Carlo Tree Search and genetic algorithm principles to produce diverse, high-quality ideas through iterative evolution and cross-domain knowledge integration.

AI-generated summary

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.

Community

Paper author Paper submitter

FlowPIE tackles scientific idea generation by combining flow-guided exploration with an evolutionary process, constructing a high-quality, diverse initial idea population and continuously improving it through selection, crossover, and cross-domain mutation. This framework overcomes reward plateaus in pure exploration and demonstrates strong test-time scaling, producing ideas with superior novelty, feasibility, and diversity compared to existing methods.

This is an automated message from the Librarian Bot. I found the following papers similar to this paper.

The following papers were recommended by the Semantic Scholar API

Please give a thumbs up to this comment if you found it helpful!

If you want recommendations for any Paper on Hugging Face checkout this Space

You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: @librarian-bot recommend

Sign up or log in to comment

Get this paper in your agent:

hf papers read 2603.29557
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2603.29557 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2603.29557 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2603.29557 in a Space README.md to link it from this page.

Collections including this paper 2