Paper summary
Paper summary: DataFlex: A Unified Framework for Data-Centric Dynamic Training of Large Language Models
DataFlex unifies dynamic data selection, domain-mixture adjustment and sample reweighting in one framework built on LLaMA-Factory, so data-centric methods can be compared fairly and dropped into normal training.
- Data-centric methods often live in separate codebases with different interfaces; DataFlex puts them behind one trainer.
- Dynamic data selection consistently beat static full-data training on MMLU for Mistral-7B and Llama-3.2-3B.