Evolving Idea Graphs with Learnable Edits-and-Commits for Multi-Agent Scientific Ideation
For researchers using LLM-empowered multi-agent systems to generate scientific ideas, EIG provides a new method to track and refine ideas through an evolving graph, improving benchmark scores and expert ratings.
EIG introduces a graph-based multi-agent framework for scientific ideation that outperforms existing systems on AI Idea Bench 2025 and LiveIdeaBench across novelty, feasibility, and clarity metrics, with explicit graph state providing main performance gains and learned edit-and-commit control adding consistent improvements.
LLM-empowered multi-agent systems offer new potential to accelerate scientific discovery by generating novel research ideas. However, existing methods typically coordinate agents through temporary texts, such as drafts or chat logs; it is difficult to pinpoint the weaknesses in the generated ideas and how the agents refine them. To this end, we introduce \textbf{Evolving Idea Graphs} (EIG), a graph-based multi-agent scientific ideation framework that can generate high-performance research ideas across various benchmark-native metrics, such as novelty, feasibility, and clarity. Instead of coordinating solely through texts, EIG represents a partially formed proposal as an evolving idea graph, where nodes capture scientific claims and edges encode relations (e.g., support and conflict), enabling unresolved weaknesses to remain identifiable throughout the idea evolving process. Specifically, a learned two-head controller operates over the evolving graph to guide the ideation: one head selects graph edits for agents to execute, while the other decides when the graph is ready for commit as final proposal synthesis. On AI Idea Bench 2025 and LiveIdeaBench, EIG outperforms all compared systems on both automatic benchmark scores and blind expert ratings. Ablations further show that explicit graph state provides the main performance gains, and learned edit-and-commit control adds consistent improvements.