Search papers, labs, and topics across Lattice.
The paper introduces RoadMap, a new benchmark for evaluating LLMs in generating research roadmaps, and identifies limitations in knowledge, task decomposition, and logical coherence. To address these limitations, they propose RoadMapper, a multi-agent system that decomposes roadmap generation into initial generation, knowledge augmentation, and iterative critique-revise-evaluate stages. Experiments show RoadMapper improves LLM performance by 8% on the RoadMap benchmark and reduces roadmap generation time by 84% compared to human experts.
LLMs can now generate research roadmaps that are not only better but also far faster than those created by human experts, thanks to a novel multi-agent system.
People commonly leverage structured content to accelerate knowledge acquisition and research problem solving. Among these, roadmaps guide researchers through hierarchical subtasks to solve complex research problems step by step. Despite progress in structured content generation, the roadmap generation task has remained unexplored. To bridge this gap, we introduce RoadMap, a novel benchmark designed to evaluate the ability of large language models (LLMs) to construct high-quality roadmaps for solving complex research problems. Based on this, we identify three limitations of LLMs: (1) lack of professional knowledge, (2) unreasonable task decomposition, and (3) disordered logical relationships. To address these challenges, we propose RoadMapper, an LLM-based multi-agent system that decomposes the research roadmap generation task into three key stages (i.e., initial generation, knowledge augmentation, and iterative "critique-revise-evaluate"). Extensive experiments demonstrate that RoadMapper can improve LLMs' ability for roadmap generation, while enhancing average performance by more than 8% and saving 84% of the time required by human experts, highlighting its effectiveness and application potential.
