Scaling Generalist Data-Analytic Agents Article Swipe

MAKE IMPACT

Shuofei Qiao , Yanqiu Zhao , Zhisong Qiu , Xiaobin Wang , Jintian Zhang , Bin Zhao , Ningyu Zhang , Yong Jiang , Pengjun Xie , Fei Huang , Huajun Chen · YOU? · · 2025 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2509.25084

Data-analytic agents are emerging as a key catalyst for automated scientific discovery and for the vision of Innovating AI. Current approaches, however, rely heavily on prompt engineering over proprietary models, while open-source models struggle to face diverse-format, large-scale data files and long-horizon, multi-step reasoning that real-world analytics demands. This paper introduces DataMind, a scalable data synthesis and agent training recipe designed to build generalist data-analytic agents. DataMind tackles three key challenges in building open-source data-analytic agents, including insufficient data resources, improper training strategy, and unstable code-based multi-turn rollout. Concretely, DataMind applies 1) a fine-grained task taxonomy and a recursive easy-to-hard task composition mechanism to increase the diversity and difficulty of synthesized queries; 2) a knowledge-augmented trajectory sampling strategy followed by model-based and rule-based filtering; 3) a dynamically adjustable training objective combining both SFT and RL losses; 4) a memory-frugal and stable code-based multi-turn rollout framework. Built on DataMind, we curate DataMind-12K, a high-quality trajectory set spanning diverse domains, task categories, and data file formats for data-analytic tasks. Trained on DataMind-12K, our DataMind-14B achieves state-of-the-art with an average score of 71.16% on multiple data analysis benchmarks, outperforming the strongest proprietary baselines DeepSeek-V3.1 and GPT-5. Our DataMind-7B also performs best among all open-source models with a score of 68.10%. We also incorporate some empirical insights gained from our exploratory trials into the analysis experiments, aiming to provide actionable insights about agentic training for the community. We will release DataMind-12K and DataMind-7B,14B for the community's future research.

Type

preprint

Language

en

Landing Page

http://arxiv.org/abs/2509.25084

PDF

https://arxiv.org/pdf/2509.25084

OA Status

green

OpenAlex ID

https://openalex.org/W4415337623

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W4415337623

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.2509.25084

Digital Object Identifier

Title

Scaling Generalist Data-Analytic Agents

Work title

Type

preprint

OpenAlex work type

Language

en

Primary language

Publication year

2025

Year of publication

Publication date

2025-09-29

Full publication date if available

Authors

Shuofei Qiao, Yanqiu Zhao, Zhisong Qiu, Xiaobin Wang, Jintian Zhang, Bin Zhao, Ningyu Zhang, Yong Jiang, Pengjun Xie, Fei Huang, Huajun Chen

List of authors in order

Landing page

https://arxiv.org/abs/2509.25084

Publisher landing page

PDF URL

https://arxiv.org/pdf/2509.25084

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/2509.25084

Direct OA link when available

Cited by

0

Total citation count in OpenAlex

Full payload