MASS: Deep Research for Social Sciences with Memory-Augmented Social Simulation
2026-06-08 • Artificial Intelligence
Artificial Intelligence
AI summaryⓘ
The authors propose a new approach called Memory-Augmented Social Simulation (MASS) to help AI models write better social science research papers. Instead of just gathering information from the internet, their method simulates realistic social interactions based on planned goals and social rules. They also use memory techniques to help the AI learn and remember important behaviors. Their tests show that this approach improves the quality and insightfulness of AI-generated research compared to existing methods.
Large Language ModelsSocial SimulationMemory-Augmented SystemsDynamic Goal-Path PlanningSocial NormsBehavior DatasetEbbinghaus Forgetting CurveAI Research WritingEmpirical FoundationInsight Generation
Authors
Yongrui Liu, Deyi Xiong
Abstract
Deep Research agents powered by Large Language Models (LLMs) have exhibited extraordinary potential in automated paper writing tasks. However, existing systems rely heavily on literature retrieval and synthesis through internet and local knowledge bases, often resulting research in lacking insight and creativity in social science. To address this issue, we propose "Memory-Augmented Social Simulation (MASS)", an innovative paradigm that leverages highly realistic and research-oriented social simulations to enhance the creativity and empirical founding of LLMs-generated research. Specifically, MASS integrates three core components: dynamic goal-path planning with multi-level social norm restraint to guide the simulation, a multi-disciplinary behavior dataset for agent memory cold-start, and a structured forgetting mechanism inspired by the Ebbinghaus curve. Together, these ensure simulation authenticity and provide a robust empirical foundation for generating innovative scholarly papers. Experimental results demonstrate the effectiveness of our method, showing a 6.81\% improvement in generation overall quality over foundation LLMs and 17.19\% gain in Insight over strong baselines.