DataEvolver: Automatic Data Preparation for Large Language Models through Multi-Level Self-Evolving

arXiv:2606.07001 · cs.DB, cs.AI · Submitted 2026-06-05 · Read on arXiv

cs.DB, cs.AI

Submitted: 2026-06-05

Updated: 2026-10-13

Code: https://github.com/ruc-datalab/DataEvolver

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Adaptability — 1× in this paper · explained in 1 episode
• Benchmark — 1× in this paper · explained in 2 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 8 episodes

Sources

• SemDeDup: Data-efficient learning at web-scale through semantic deduplication
• Prompt2DAG: A Modular Methodology for LLM-Based Data Enrichment Pipeline Generation
• DoPAMine: Domain-specific Pre-training Adaptation from seed-guided data Mining
• LLaPipe: LLM-Guided Reinforcement Learning for Automated Data Preparation Pipeline Construction
• Data-Juicer: A One-Stop Data Processing System for Large Language Models
• Data-Juicer 2.0: Cloud-Scale Adaptive Data Processing for and with Foundation Models
• AlpaGasus: Training A Better Alpaca with Fewer Data
• Teaching Large Language Models to Self-Debug
• Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
• Training Verifiers to Solve Math Word Problems
• MoDS: Model-oriented Data Selection for Instruction Tuning
• Length-Controlled AlpacaEval: A Simple Way to Debias Automatic Evaluators
• AutoPrep: Natural Language Question-Aware Data Preparation with a Multi-Agent Framework
• A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems
• Text-to-Pipeline: Bridging Natural Language and Data Preparation Pipelines
• Gemma 3 Technical Report
• The Llama 3 Herd of Models
• Measuring Mathematical Problem Solving With the MATH Dataset
• Training Compute-Optimal Large Language Models
• MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework

Related papers

• Cost-Aware Mixture-of-Experts Coordination for Model Markets
• Vibe Coding on Trial: Operating Characteristics of Unanimous LLM Juries
• Human-Level Text-to-SQL via Reinforcement Learning on Verified Data, Without Pipeline Engineering
• MetaJoin: Adaptive Join Selections for Heterogeneous Tables
• Bridging Business Intent and Data: A Benchmark for Automatic Relational Data Product Generation
• DexterSQL: Deep Schema Exploration and Rule-based Correction for Text-to-SQL Generation