Scaling Parameter-Constrained Language Models with Quality Data
원문 초록
Scaling laws in language modeling traditionally quantify training loss as a function of dataset size and model parameters, providing compute-optimal estimates but often neglecting the impact of data quality on model generalization. In this paper, we extend the conventional understanding of scaling law by offering a microscopic view of data quality within the original formulation -- effective training tokens -- which we posit to be a critical determinant of performance for parameter-constrained language models. Specifically, we formulate the proposed term of effective training tokens to be a combination of two readily-computed indicators of text: (i) text diversity and (ii) syntheticity as measured by a teacher model. We pretrained over $200$ models of 25M to 1.5B parameters on a diverse set of sampled, synthetic data, and estimated the constants that relate text quality, model size, training tokens, and eight reasoning task accuracy scores. We demonstrated the estimated constants yield +0.83 Pearson correlation with true accuracies, and analyzed it in scenarios involving widely-used data techniques such as data sampling and synthesis which aim to improve data quality.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
Intent Factored Generation: Unleashing the Diversity in Your Language Model
Mixture of Experts in Large Language Models
AdapThink: Adaptive Thinking Preferences for Reasoning Language Model
Unsourced Adversarial CAPTCHA: A Bi-Phase Adversarial CAPTCHA Framework
Mlphon: A Multifunctional Grapheme-Phoneme Conversion Tool Using Finite State Transducers
Specification-based Security Evaluation of Real-World Agents Tool-Calling under Long-Horizon Context Tasks
토론
이 논문과 연결된 포럼 글이 아직 없습니다.