Multi-agent framework for high-quality scientific survey generation using a reflect-when-you-write paradigm. Features six specialized agents (Interpreter, Organizer, Collector, Composer, Refiner, Reflector) with a hierarchical Reflector that evaluates drafts at outline, section, and document levels. Outperforms LLM-x-MapReduce-V2 and AutoSurvey by +1.73 in coherence and +32% citation F1. Includes SurveyScope benchmark of 46 high-impact papers (2020-2025) across 11 CS domains.

Paper

arXiv: 2506.12689

agentsscientific-researchopen-source