Elamur Research⌕
☰
home / papers / 2026 / MASS: Deep Research for Social Sciences with Memory-Augmented Social Simulation

MASS: Deep Research for Social Sciences with Memory-Augmented Social Simulation

Paper · 2026 · Yongrui Liu and Deyi Xiong

Summary

LLM-based Deep Research agents write papers mostly by retrieving and recombining existing literature. For the social sciences this tends to give shallow, unoriginal work, because insight there depends on confronting theory with observed behavior. The authors propose MASS (Memory-Augmented Social Simulation), which adds a simulated society to the research pipeline so the agent can produce its own empirical data. The pipeline has four stages:

  • a Stepwise Best-of-N (SW-BoN) planner that scores and votes on candidate reasoning steps;
  • tool use (web search, academic databases, and the simulator itself);
  • simulation experiments designed under the ODD protocol, with dynamic goal planning and layered social-norm constraints;
  • template-guided academic writing.

Simulated agents start with memories drawn from a newly built multidisciplinary social-behavior dataset. They also forget over time according to an Ebbinghaus-style curve, which keeps long contexts manageable and behavior realistic. Papers generated by MASS score higher on overall quality than those from foundation LLMs and show markedly more insight than strong baselines. This points to simulation-grounded evidence as a promising route to more creative automated social research.

Key points

  • Reported gains: +6.81% overall generation quality versus foundation LLMs, and +17.19% on Insight versus strong baselines.
  • SW-BoN creates n candidate nodes at each step and filters them by a 0–10 score threshold. It then chooses a winner through multi-agent consensus based on semantic similarity, with the goal of reducing error accumulation.
  • Dataset quality was checked with DeepEval:
    • 94.21% of samples pass hallucination detection and 97.51% pass integrity checks.
    • Information density is 0.89 and textual richness is 0.75.
  • Cold-starting agents from the dataset improves their action rationality score by 5.84%.
  • Sensitivity tests:
    • Small (±10%) shifts in the forgetting probability barely change results, while extreme settings hurt fidelity and speed.
    • Code-specialized LLMs produce more complete agent attributes than general-purpose models.
  • The authors say code and data will be released at github.com/tjunlp-lab/MASS_DeepResearch. They list limitations of high compute cost and hallucinations, both in generated text and in agent behavior.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓Tianjin UniversityTianjin · China52 works · 1,128 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksArtificial Intelligence · 17,965 worksArts and Humanities · 17,450 worksLanguage and Linguistics · 17,296 works

Cite

Cite as Yongrui Liu and Deyi Xiong. 2026. MASS: Deep Research for Social Sciences with Memory-Augmented Social Simulation. Findings of the Association for Computational Linguistics: ACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.findings-acl.988

@inproceedings{liu2026mass,
  title = {{MASS: Deep Research for Social Sciences with Memory-Augmented Social Simulation}},
  author = {Yongrui Liu and Deyi Xiong},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.988},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.988}
}

References

43 references, 6 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Scott Ashworth, Christopher R Berry, and Ethan Bueno De Mesquita. 2021. Theory and Credibility: Integrating Theoretical and Empirical Social Science. Princeton University Press. find
  2. Howard S Becker. 2008. Writing for Social Scientists: How to Start and Finish Your Thesis, Book, or Article. University of Chicago Press. find
  3. China SME Development Promotion Center. 2024. 2024 Annual Assessment Report on the Development Environment of Small and Medium-Sized Enterprises. Technical report, China SME Development Promotion Center. find
  4. Guoxin Chen, Zile Qiao, Wenqing Wang, Donglei Yu, Xuanzhong Chen, Hao Sun, Minpeng Liao, Kai Fan, Yong Jiang, Pengjun Xie, Wayne Xin Zhao, Ruihua Song, and Fei Huang. 2025. MARS: Optimizing Dual-System Deep Research via Multi-Agent Reinforcement Learning. arXiv preprint arXiv:2510.04935. — [arXiv:2510.04935](https://arxiv.org/abs/2510.04935) find
  5. Mingxuan Du, Benfeng Xu, Chiwei Zhu, Xiaorui Wang, and Zhendong Mao. 2025. DeepResearch Bench: A Comprehensive Benchmark for Deep Research Agents. Preprint, arXiv:2506.11763.
  6. Chen Gao, Xiaochong Lan, Zhihong Lu, Jinzhu Mao, Jinghua Piao, Huandong Wang, Depeng Jin, and Yong Li. 2023. S3: Social-Network Simulation System with Large Language Model-Empowered Agents. arXiv preprint arXiv:2307.14984. — [arXiv:2307.14984](https://arxiv.org/abs/2307.14984) find
  7. Klarita Gërxhani, Nan Dirk De Graaf, and Werner Raub. 2022. Handbook of Sociological Science. Cheltenham: Edward Elgar. find
  8. Volker Grimm, Uta Berger, Finn Bastiansen, Sigrunn Eliassen, Vincent Ginot, Jarl Giske, John Goss-Custard, Tamara Grand, Simone K Heinz, Geir Huse, and 1 others. 2006. A Standard Protocol for Describing Individual-based and Agent-based Models. Ecological modelling, 198(1-2):115–126. find
  9. Volker Grimm, Steven F Railsback, Christian E Vin-cenot, Uta Berger, Cara Gallagher, Donald L DeAngelis, Bruce Edmonds, Jiaqi Ge, Jarl Giske, Juergen Groeneveld, and 1 others. 2020. The ODD Protocol for Describing Agent-based and Other Simulation Models: A Second Update to Improve Clarity, Replicati… find
  10. Zishan Guo, Yufei Huang, and Deyi Xiong. 2024. CToolEval: A Chinese Benchmark for LLM-Powered Agent Evaluation in Real-World API Interactions. In Findings of the Association for Computational Linguistics: ACL 2024, pages 15711–15724, Bangkok, Thailand. Association for Computational Linguistics. find
  11. Zishan Guo, Renren Jin, Chuang Liu, Yufei Huang, Dan Shi, Supryadi, Linhao Yu, Yan Liu, Jiaxuan Li, Bojian Xiong, and Deyi Xiong. 2023. Evaluating Large Language Models: A Comprehensive Survey. Preprint, arXiv:2310.19736.
  12. Zhongyuan Han. 2024. The Technology of Conflict and Endogenous Authority: The Problem of the Origin of the State. Economic Theory and Business Management, 44(10):148–164. find
  13. Jeffrey Ip, Cale, and contributors. 2024. Deepeval: The LLM Evaluation Framework. find
  14. Gaurav Koley. 2025. SALM: A Multi-Agent Framework for Language Model-Driven Social Network Simulation. arXiv preprint arXiv:2505.09081. — [arXiv:2505.09081](https://arxiv.org/abs/2505.09081) find
  15. Yu Lei, Shuzheng Si, Wei Wang, Yifei Wu, Gang Chen, Fanchao Qi, and Maosong Sun. 2025. Rhinoinsight: Improving Deep Research through Control Mechanisms for Model Behavior and Context. arXiv preprint arXiv:2511.18743. — [arXiv:2511.18743](https://arxiv.org/abs/2511.18743) find
  16. Yongqi Leng, Yikun Lei, Xikai Liu, Meizhi Zhong, Bojian Xiong, Yurong Zhang, Yan Gao, Yiwu, Yao Hu, and Deyi Xiong. 2025. DecEx-RAG: Boosting Agentic Retrieval-Augmented Generation with Decision and Execution Optimization via Process Supervision. In Proceedings of the 2025 Conference on Empirical Me… find
  17. Shen Li, Yong Zhang, Yu-Xiang Wang, and An Bo. 2025a. MF-LLM: Simulating Population Decision Dynamics via A Mean-Field Large Language Model Framework. In Advances in Neural Information Processing Systems 38 (NeurIPS). find
  18. Zhigen Li, Jianxiang Peng, Yanmeng Wang, Yong Cao, Tianhao Shen, Minghui Zhang, Linxi Su, Shang Wu, Yihang Wu, YuQian Wang, Ye Wang, Wei Hu, Jianfeng Li, Shaojun Wang, Jing Xiao, and Deyi Xiong. 2025b. ChatSOP: An SOP-Guided MCTS Planning Framework for Controllable LLM Dialogue Agents. In Proceeding… find
  19. Chuang Liu, Renren Jin, Yuqi Ren, and Deyi Xiong. 2024. LHMKE: A Large-scale Holistic Multi-subject Knowledge Evaluation Benchmark for Chinese Large Language Models. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING… find
  20. Yusong Ma, Hongxuan Nie, Chao Chen, Jiujie Zhang, Jiali Jiang, Bisheng Wang, and Yuqin Xia. 2025. A Survey of Retrieval-Augmented Generation (RAG) for Large Language Models. In 2025 International Conference on Trustworthy Big Data and Artificial Intelligence (ICTBAI), pages 7–13. find
  21. Benjamin S Manning, Kehang Zhu, and John J Horton. 2024. Automated Social Science: Language Models as Scientist and Subjects. Technical report, National Bureau of Economic Research. find
  22. Scott Merrill and Shashank Srivastava. 2025. Point of Order: Action-Aware LLM Persona Modeling for Realistic Civic Simulation. arXiv preprint arXiv:2511.17813. — [arXiv:2511.17813](https://arxiv.org/abs/2511.17813) find
  23. Jaap MJ Murre and Joeri Dros. 2015. Replication and Analysis of Ebbinghaus’ Forgetting Curve. PloS one, 10(7):e0120644. find
  24. Reiichiro Nakano, Jacob Hilton, Suchir Balaji, Jeff Wu, Long Ouyang, Christina Kim, Christopher Hesse, Shantanu Jain, Vineet Kosaraju, William Saunders, and 1 others. 2021. Webgpt: Browser-assisted Question-answering with Human Feedback. arXiv preprint arXiv:2112.09332.
  25. Jinghua Piao, Yuwei Yan, Jun Zhang, Nian Li, Junbo Yan, Xiaochong Lan, Zhihong Lu, Zhiheng Zheng, Jing Yi Wang, Di Zhou, and 1 others. 2025. Agentsociety: Large-scale Simulation of Llm-driven Generative Agents Advances Understanding of Human Behaviors and Society. arXiv preprint arXiv:2502.08691. — [arXiv:2502.08691](https://arxiv.org/abs/2502.08691) find
  26. Riwa Saied. 2025. How to Use Perplexity AI (2025): Screenshots, Citations & Setup. Online Technical Guide. find
  27. Jiabin Tang, Tianyu Fan, and Chao Huang. 2025. Autoagent: A Fully-Automated and Zero-Code Framework for LLM Agents. arXiv preprint arXiv:2502.05957.
  28. Tongyi DeepResearch Team, Baixuan Li, Bo Zhang, Dingchu Zhang, Fei Huang, Guangyu Li, Guoxin Chen, Huifeng Yin, Jialong Wu, Jingren Zhou, and 1 others. 2025. Tongyi Deepresearch Technical Report. arXiv preprint arXiv:2510.24701. — [arXiv:2510.24701](https://arxiv.org/abs/2510.24701) find
  29. Tuong Manh Vu, Charlotte Probst, Alexandra Nielsen, Hao Bai, Charlotte Buckley, Petra S Meier, Mark Strong, Alan Brennan, and Robin C Purshouse. 2020. A Software Architecture for Mechanism-Based Social Systems Modelling in Agent-Based Simulation Models. Journal of artificial societies and social sim… find
  30. Yi Wan, Jiuqi Wang, Liam Li, Jinsong Liu, Ruihao Zhu, and Zheqing Zhu. 2025. Pokeeresearch: Effective Deep Research via Reinforcement Learning from AI Feedback and Robust Reasoning Scaffold. arXiv preprint arXiv:2510.15862. — [arXiv:2510.15862](https://arxiv.org/abs/2510.15862) find
  31. Lei Wang, Heyang Gao, Xiaohe Bo, Xu Chen, and Ji-Rong Wen. 2025. Yulan-onesim: Towards the Next Generation of Social Simulator with Large Language Models. In Workshop on Scaling Environments for Agents. find
  32. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, and 1 others. 2022. Chain-of-Thought Prompting Elicits Rreasoning in Large Language Models. volume 35, pages 24824–24837. find
  33. John T Wixted. 2022. The Enigma of Forgetting. Proceedings of the National Academy of Sciences, 119(12):e2201332119. find
  34. Zihao Wu. 2025. Autono: A React-Based Highly Robust Autonomous Agent Framework. arXiv preprint arXiv:2504.04650. — [arXiv:2504.04650](https://arxiv.org/abs/2504.04650) find
  35. Bojian Xiong, Yikun Lei, Xikai Liu, Shaowei Zhang, Pengyun Zhu, Yan Liu, Yongqi Leng, Ling Shi, Meizhi Zhong, Yurong Zhang, Yan Gao, Yiwu, Yao Hu, and Deyi Xiong. 2025. Think-Search-Patch: A Retrieval-Augmented Reasoning Framework for Repository-Level Code Repair. In Proceedings of the 2025 Conferen… find
  36. Renjun Xu and Jingwen Peng. 2025. A Comprehensive Survey of Deep Research: Systems, Methodologies, and Applications. arXiv preprint arXiv:2506.12594. — [arXiv:2506.12594](https://arxiv.org/abs/2506.12594) find
  37. Xiao Xue, Deyu Zhou, Ming Zhang, and Fei-Yue Wang. 2025. From Agent Simulation to Social Simulator: A Comprehensive Review (Part 1). arXiv preprint arXiv:2510.18271. — [arXiv:2510.18271](https://arxiv.org/abs/2510.18271) find
  38. Shunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran, Tom Griffiths, Yuan Cao, and Karthik Narasimhan. 2023. Tree of Thoughts: Deliberate Problem Solving with Large Language Models. In Advances in Neural Information Processing Systems, volume 36, pages 11809–11822. Curran Associates, Inc. find
  39. Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik R Narasimhan, and Yuan Cao. 2022. React: Synergizing Reasoning and Acting in Language Models. In The eleventh international conference on learning representations.
  40. Shiwei Zhang, Lansong Diao, Siyu Wang, Zongyan Cao, Yiliang Gu, Chang Si, Ziji Shi, Zhen Zheng, Chuan Wu, and Wei Lin. 2023. Auto-parallelizing Large Models with Rhino: A Systematic Approach on Production AI Platform. arXiv preprint arXiv:2302.08141. — [arXiv:2302.08141](https://arxiv.org/abs/2302.08141) find
  41. Wenlin Zhang, Xiaopeng Li, Yingyi Zhang, Pengyue Jia, Yichao Wang, Huifeng Guo, Yong Liu, and Xiangyu Zhao. 2025. Deep Research: A Survey of Autonomous Research Agents. arXiv preprint arXiv:2508.12752. — [arXiv:2508.12752](https://arxiv.org/abs/2508.12752) find
  42. Yuxiang Zheng, Dayuan Fu, Xiangkun Hu, Xiaojie Cai, Lyumanshan Ye, Pengrui Lu, and Pengfei Liu. 2025. DeepResearcher: Scaling Deep Research via Reinforcement Learning in Real-world Environments. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 414– 431…
  43. King Zhu, Hanhao Li, Siwei Wu, Tianshun Xing, Dehua Ma, Xiangru Tang, Minghao Liu, Jian Yang, Jiaheng Liu, Yuchen Eleanor Jiang, and 1 others. 2025. Scaling Test-time Compute for LLM Agents. arXiv preprint arXiv:2506.12928. — [arXiv:2506.12928](https://arxiv.org/abs/2506.12928) find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/liu-2026-mass-deep-research-social-sciences/