Scaling Laws of Synthetic Data for Language Models
원문 초록
Large language models (LLMs) achieve strong performance across diverse tasks, largely driven by high-quality web data used in pre-training. However, recent studies indicate this data source is rapidly depleting. Synthetic data emerges as a promising alternative, but it remains unclear whether synthetic datasets exhibit predictable scalability comparable to raw pre-training data. In this work, we systematically investigate the scaling laws of synthetic data by introducing SynthLLM, a scalable framework that transforms pre-training corpora into diverse, high-quality synthetic datasets. Our approach achieves this by automatically extracting and recombining high-level concepts across multiple documents using a graph algorithm. Key findings from our extensive mathematical experiments on SynthLLM include: (1) SynthLLM generates synthetic data that reliably adheres to the rectified scaling law across various model sizes; (2) Performance improvements plateau near 300B tokens; and (3) Larger models approach optimal performance with fewer training tokens. For instance, an 8B model peaks at 1T tokens, while a 3B model requires 4T. Moreover, comparisons with existing synthetic data generation and augmentation methods demonstrate that SynthLLM achieves superior performance and scalability. Our findings highlight synthetic data as a scalable and reliable alternative to organic pre-training corpora, offering a viable path toward continued improvement in model performance.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
How Good Are Synthetic Requirements ? Evaluating LLM-Generated Datasets for AI4RE
SDialog: A Python Toolkit for Synthetic Dialogue Generation and Analysis
Cosmos-Drive-Dreams: Scalable Synthetic Driving Data Generation with World Foundation Models
Optimization-Free Universal Watermark Forgery with Regenerative Diffusion Models
BEAR: BGP Event Analysis and Reporting
Datasheets Aren't Enough: DataRubrics for Automated Quality Metrics and Accountability
토론
이 논문과 연결된 포럼 글이 아직 없습니다.