Enhancing Research Idea Generation through Combinatorial Innovation and Multi-Agent Iterative Search Strategies

2026-04-22

AI Summary

The exponential growth of scientific literature has substantially increased the cost of identifying high-quality research ideas, while existing large language models often produce repetitive and superficial proposals. To address this challenge, this work proposes a multi-agent iterative planning and search framework inspired by combinatorial innovation theory—the first to integrate this theoretical foundation with LLM-driven multi-agent systems. By iteratively generating, retrieving relevant knowledge, evaluating, and refining research ideas, the framework progressively enhances both novelty and diversity. Experimental results in the domain of natural language processing demonstrate that the proposed approach significantly outperforms current baselines, yielding idea quality comparable to that of papers accepted at top-tier conferences, yet distinguishable from clearly rejected submissions.

Technology Category

• Multiagent Systems: Multiagent Planning
• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Scientific progress depends on the continual generation of innovative re-search ideas. However, the rapid growth of scientific literature has greatly increased the cost of knowledge filtering, making it harder for researchers to identify novel directions. Although existing large language model (LLM)-based methods show promise in research idea generation, the ideas they produce are often repetitive and lack depth. To address this issue, this study proposes a multi-agent iterative planning search strategy inspired by com-binatorial innovation theory. The framework combines iterative knowledge search with an LLM-based multi-agent system to generate, evaluate, and re-fine research ideas through repeated interaction, with the goal of improving idea diversity and novelty. Experiments in the natural language processing domain show that the proposed method outperforms state-of-the-art base-lines in both diversity and novelty. Further comparison with ideas derived from top-tier machine learning conference papers indicates that the quality of the generated ideas falls between that of accepted and rejected papers. These results suggest that the proposed framework is a promising approach for supporting high-quality research idea generation. The source code and dataset used in this paper are publicly available on Github repository: https://github.com/ChenShuai00/MAGenIdeas. The demo is available at https://huggingface.co/spaces/cshuai20/MAGenIdeas.

Problem

• research idea generation
• combinatorial innovation
• large language models
• idea novelty
• knowledge filtering

Innovation

• combinatorial innovation
• multi-agent system
• iterative search
• research idea generation
• large language models