IDEAgent: Agentic Quality-Diversity Search for Research Idea Generation

2026-07-24

AI Summary

Existing approaches to research idea generation often optimize either quality or diversity in isolation, resulting in outputs that are either homogeneous or lacking in rigor and clarity. This work addresses this limitation by introducing, for the first time, a joint quality-diversity (QD) search framework for this task, proposing IDEAgent—a multi-agent system that synergistically enhances both dimensions through a lineage-based evolutionary mechanism. Key innovations include the Yield metric for unified QD evaluation, integrated with lightweight sequential memory, explicit historical comparison, and a multi-objective feedback-driven refinement process. Experiments across 32 computer science topics demonstrate that Yield improves by 3.89× over the strongest baseline and achieves non-zero Yield on eight times as many topics, substantially boosting the logical rigor, clarity, and novelty of generated ideas.

Abstract

Large Language Models (LLMs) have significantly automated the process of scientific discovery over the past few years. However, existing systems share one core limitation: they generate and optimize ideas independently for either Quality or Diversity. This often leads to the generation of ideas in close proximity to one another or to a large set of trivial, unsound, or unclear concepts. In this work, we instead argue that research ideation should be treated as a conjunction of both objectives and framed as a Quality-Diversity (QD) search. In line with this perspective, we introduce IDEAgent, a multi-agent framework that manages the evolution of ideas through lineages. We jointly drive Quality using multi-objective feedback for dedicated repair and refinement, while Diversity is achieved through lightweight sequential memory and explicit comparison against completed ideas, their historical ancestors, and rejected proposals. To systematically evaluate this QD conjunction, we develop Yield, a joint metric that computes the largest set of mutually diverse ideas that satisfy a predetermined quality threshold. Finally, through evaluations across 32 topics spanning 8 domains of Computer Science, we show that IDEAgent outperforms the best baseline by 3.89x on Yield, while achieving non-zero Yield on 8x more topics. We further corroborate these findings through an analysis of quality improvements, showing that repair and refinement are crucial for building logical rigor and clarity while preserving non-obviousness. To encourage future research on QD-search-based ideation, we open-source IDEAgent at https://github.com/declare-lab/IDEAgent.

Problem

• Quality-Diversity Search
• Research Idea Generation
• Large Language Models
• Scientific Discovery
• Idea Optimization

Innovation

• Quality-Diversity Search
• Multi-agent Framework
• Research Idea Generation
• LLM-based Ideation
• Yield Metric