Large-Scale Diverse Synthesis for Mid-Training
This addresses the problem of limited knowledge diversity and scalability in training data for LLM developers, though it is incremental as it builds on existing synthesis methods.
The paper tackles the scarcity of high-quality training data for large language models by proposing a novel pipeline to synthesize BoostQA, a 100B-token QA dataset, which when used in mid-training improves Llama-3 8B by 12.74% on MMLU and CMMLU and achieves state-of-the-art performance across 12 benchmarks.
The scarcity of high-quality, knowledge-intensive training data hinders the development of large language models (LLMs), as traditional corpora provide limited information. Previous studies have synthesized and integrated corpora-dependent question-answering (QA) data to improve model performance but face challenges in QA data scalability and knowledge diversity, particularly in cross-domain contexts. Furthermore, leveraging our designed discipline and difficulty annotation system, we probe model deficiencies in STEM disciplines and high-difficulty data. To overcome these limitations, we propose a novel diversified pipeline to synthesize BoostQA, a 100B-token large-scale QA dataset. Our synthesis framework: (1) curates seed data from heterogeneous sources; (2) utilizes DeepSeek-R1 to implement STEM-focused multi-grade synthesis to boost data diversity and high-difficulty synthesis to mitigate difficulty degradation; (3) refines answers via DeepSeek-V3 to improve output quality. We utilize BoostQA in mid-training, a mid-stage between pre-training and post-training, to optimize domain-specific knowledge acquisition and enhance data quality. Our method enables Llama-3 8B, mid-trained on a 40B-token dataset, to achieve an average improvement of $\mathbf{12.74\%}$ on MMLU and CMMLU and establish SOTA average performance across 12 benchmarks. BoostQA also demonstrates robust scalability, with performance consistently improving as model size, data volume, and initial FLOPs scale.