The Year in AI Papers
The year/Independent research

Paper 2512.16676

DataFlow: An LLM-Driven Framework for Unified Data Preparation and Workflow Automation in the Era of Data-Centric AI

Hao Liang, Xiaochen Ma, Zhou Liu, Zhen Hao Wong, Zhengyang Zhao, Zimo Meng, Runming He, Chengyu Shen, Qifeng Cai, Zhaoyang Han, Meiyi Qiang, Yalin Feng, et al.

Published
Dec 2025
Research lab
Independent
Citations
19
GitHub
7.2K stars

01 In brief

Summary

DataFlow is a unified, LLM-driven framework for data preparation in the era of data-centric AI, addressing the fragmentation of ad-hoc scripts and workflows.

It provides a PyTorch-style API with nearly 200 reusable operators and six domain-general pipelines (text, math, code, Text-to-SQL, agentic RAG, knowledge extraction).

The framework includes a global storage abstraction, hierarchical programming interfaces, and a DataFlow-Agent that translates natural-language specifications into executable pipelines via operator synthesis and iterative verification.

Experiments show DataFlow consistently improves downstream LLM performance: math pipelines achieve 1–3 point gains on MATH, GSM8K, and AIME; Text-to-SQL pipelines improve execution accuracy by over 3% compared to SynSQL; code pipelines deliver over 7% average improvements.

A unified 10K-sample dataset (DataFlow-Instruct-10K) enables base models to surpass counterparts trained on 1M Infinity-Instruct data.

The framework is open-source, with code and datasets released, and aims to establish a system-level foundation for reproducible, scalable LLM data preparation.

02 From the paper

Abstract

The rapidly growing demand for high-quality data in Large Language Models (LLMs) has intensified the need for scalable, reliable, and semantically rich data preparation pipelines. However, current practices remain dominated by ad-hoc scripts and loosely specified workflows, which lack principled abstractions, hinder reproducibility, and offer limited support for model-in-the-loop data generation. To address these challenges, we present DataFlow, a unified and extensible LLM-driven data preparation framework. DataFlow is designed with system-level abstractions that enable modular, reusable, and composable data transformations, and provides a PyTorch-style pipeline construction API for building debuggable and optimizable dataflows. The framework consists of nearly 200 reusable operators and six domain-general pipelines spanning text, mathematical reasoning, code, Text-to-SQL, agentic RAG, and large-scale knowledge extraction. To further improve usability, we introduce DataFlow-Agent, which automatically translates natural-language specifications into executable pipelines via operator synthesis, pipeline planning, and iterative verification. Across six representative use cases, DataFlow consistently improves downstream LLM performance. Our math, code, and text pipelines outperform curated human datasets and specialized synthetic baselines, achieving up to +3\% execution accuracy in Text-to-SQL over SynSQL, +7\% average improvements on code benchmarks, and 1--3 point gains on MATH, GSM8K, and AIME. Moreover, a unified 10K-sample dataset produced by DataFlow enables base models to surpass counterparts trained on 1M Infinity-Instruct data. These results demonstrate that DataFlow provides a practical and high-performance substrate for reliable, reproducible, and scalable LLM data preparation, and establishes a system-level foundation for future data-centric AI development.