@inproceedings{liu-etal-2026-dataseer,
    title = "{D}ata{S}eer: A Manager-Centric Collaborative Multi-Agent Framework with Multi-Branch Reasoning for Automated Insight Discovery",
    author = "Liu, Suchen  and
      Song, Yuanfeng  and
      Gao, Jun  and
      Chen, Xing",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.findings-acl.787/",
    pages = "16030--16058",
    ISBN = "979-8-89176-395-1",
    abstract = "The growth of complex data fuels demand for automated insight discovery. While LLMs and agent technologies have advanced data analysis, existing methods struggle with maintaining contextual coherence, limited coverage due to single-path exploration, and rigid planning that fails to adapt to dynamic data discovery. We propose DataSeer, a collaborative multi-agent framework for automated insight discovery. Our first contribution is a \textbf{Manager-Centric Collaborative Framework}, where the Manager ensures cross-episode contextual coherence through a dual-layer memory system with compression, consolidation, and retrieval, alongside dynamic prompt editing, coordinating the overall process between the Planner and Executor. Second, we \textbf{optimize the planning and execution components}: the Planner employs multi-role discussion for adaptive sub-goal generation and plan refinement; the Executor is endowed with tactical autonomy for exploratory execution and incorporates real-time multi-dimensional self-assessment to guarantee insight quality. Third, we design \textbf{Multi-Branch Reasoning} that executes multiple discovery trajectories and synthesizes outcomes through LLM-based aggregation, improving coverage and reducing single-path stochasticity. Experiments on InsightBench and InsightEval show that DataSeer outperforms baselines, achieving improvements of 18.7{\%} and 12.1{\%} in insight-level scores, and 11.6{\%} and 10.3{\%} in summary-level scores, respectively."
}