Large Language Model-Driven Data Analysis Agents: A Systematic Review of Architectures and Code Generation Strategies
DOI:
https://doi.org/10.54097/9349dq61Keywords:
Large Language Model, Data Analysis Agent, Multi-agent System, Text-to-SQL, Code Generation, Autonomous Data ScienceAbstract
Autonomous data analysis, once the domain of experienced programmers, is being reshaped by large language models (LLMs) — a shift this paper examines through a structured systematic review. Models like GPT-4, Claude, and LLaMA have reached a level of proficiency in natural language understanding and code generation that now makes it feasible to embed them inside agent-based pipelines capable of planning, executing, and interpreting analytical tasks with little human intervention. To bring conceptual order to this rapidly growing body of work, the review proposes a two-dimensional taxonomy organized around two design choices: system architecture (single-agent versus multi-agent) and code generation modality (Python-centric versus SQL-centric). Within this framework, three lines of research emerge as dominant: systems built around a single execution agent, multi-agent collaborative pipelines, and text-to-SQL generation architectures. Drawing on representative work published at leading venues between 2023 and 2025, each line is assessed with respect to planning strategies, execution feedback loops, and error-correction mechanisms. A cross-cutting analysis finds that how a system handles execution feedback — rather than whether it uses one agent or several — is the factor that most reliably separates stronger from weaker designs. The paper closes by mapping out three underexplored problems: verifying code across modalities, developing unified evaluation benchmarks, and extending multi-agent coordination into SQL-oriented settings.
Downloads
References
[1] Wang, L., Ma, C., Feng, X., et al. (2024). A survey on large language model based autonomous agents. Frontiers of Computer Science. DOI: https://doi.org/10.1007/s11704-024-40231-1
[2] Li, G., Zhou, X., & Zhao, X. (2024). LLM for data management. Proceedings of the VLDB Endowment, 17(12), 4213–4216. DOI: https://doi.org/10.14778/3685800.3685838
[3] Sun, M., Han, R., Jiang, B., et al. (2025). A survey on LLM-based agents for statistics and data science. The American Statistician. DOI: https://doi.org/10.1080/00031305.2025.2561140
[4] Luo, Y., et al. (2026). Data agents: Levels, state of the art, and open problems. SIGMOD Tutorial. DOI: https://doi.org/10.1145/3788853.3801878
[5] Hong, S., Lin, Y., Liu, B., et al. (2025). Data interpreter: An LLM agent for data science. In ACL Findings. DOI: https://doi.org/10.18653/v1/2025.findings-acl.1016
[6] Guo, S., Deng, C., Wen, Y., et al. (2024). DS-Agent: Automated data science by empowering LLMs with case-based reasoning. In ICML.
[7] Qiao, B., Li, L., Zhang, X., et al. (2023). TaskWeaver: A code-first agent framework. arXiv:2311.17541.
[8] Pourreza, M., & Rafiei, D. (2023). DIN-SQL: Decomposed in-context learning of text-to-SQL with self-correction. In NeurIPS. DOI: https://doi.org/10.52202/075280-1577
[9] Pourreza, M., Li, H., Sun, R., et al. (2025). CHASE-SQL: Multi-path reasoning and preference optimized candidate selection in text-to-SQL. In ICLR.
[10] Hong, S., Zheng, X., Chen, J., et al. (2024). MetaGPT: Meta programming for multi-agent collaborative framework. In ICLR.
[11] Sun, M., Han, R., Jiang, B., et al. (2025). LAMBDA: A large model based data agent. Journal of the American Statistical Association. DOI: https://doi.org/10.1080/01621459.2025.2510000
[12] Yang, Z., Zhou, Z., Wang, S., et al. (2024). MatPlotAgent: Method and evaluation for LLM-based agentic scientific data visualization. In ACL Findings. DOI: https://doi.org/10.18653/v1/2024.findings-acl.701
[13] Wang, B., Ren, C., Yang, J., et al. (2025). MAC-SQL: A multi-agent collaborative framework for text-to-SQL. In COLING.
[14] Huang, Y., Luo, J., Yu, Y., et al. (2024). DA-Code: Agent data science code generation benchmark for LLMs. In EMNLP. DOI: https://doi.org/10.18653/v1/2024.emnlp-main.748
[15] Li, J., Hui, B., Qu, G., et al. (2023). Can LLM already serve as a database interface? A big bench for large-scale database grounded text-to-SQLs. In NeurIPS. DOI: https://doi.org/10.52202/075280-1835
[16] Gao, D., Wang, H., Li, Y., et al. (2024). Text-to-SQL empowered by large language models: A benchmark evaluation. Proceedings of the VLDB Endowment, 17(5), 1132–1145. DOI: https://doi.org/10.14778/3641204.3641221
[17] Hong, Z., Yuan, Z., Zhang, Q., et al. (2025). Next-generation database interfaces: A survey of LLM-based text-to-SQL. IEEE Transactions on Knowledge and Data Engineering. DOI: https://doi.org/10.1109/TKDE.2025.3609486
[18] Jiang, J., Xie, H., Shen, S., et al. (2025). SiriusBI: A comprehensive LLM-powered solution for data analytics in business intelligence. Proceedings of the VLDB Endowment, 18(12), 4860–4873. DOI: https://doi.org/10.14778/3750601.3750610
[19] Li, P., He, Y., Yashar, D., et al. (2024). Table-GPT: Table fine-tuned GPT for diverse table tasks. Proceedings of the ACM on Management of Data, 2(3), 176. DOI: https://doi.org/10.1145/3654979
[20] Grosnit, A., Maraval, A., et al. (2024). Large language models orchestrating structured reasoning achieve Kaggle grandmaster level. arXiv:2411.03562.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Frontiers in Computing and Intelligent Systems
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
