arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2502.02494cs.LGcs.CL
收藏

面向语言模型预训练数据选择的相似性度量分析

Analyzing Similarity Metrics for Data Selection for Language Model Pretraining

Carnegie Mellon University(卡内基梅隆大学);Google Research(谷歌研究)

Dylan Sam, Ayan Chakrabarti, Afshin Rostamizadeh, Srikumar Ramalingam, Gui Citovsky, Sanjiv Kumar

AI总结:

该研究提出评估框架验证了通用现成嵌入模型不适配预训练数据构建场景,其表现弱于同语料训练模型提取的极简嵌入,为预训练专用嵌入设计奠定基础。

摘要:

衡量训练样本间的相似性对构建高质量、多样化的语言模型预训练数据集至关重要。但目前相似性通常采用为检索等任务训练的通用现成嵌入模型计算,这类基于嵌入的相似性度量是否适用于预训练数据选择仍有待深入探究。\n本文提出一个新框架,专门评估相似性度量在语言模型预训练数据构建场景中的适用性。框架的第一个评估标准衡量距离指标能在多大程度上反映不同训练样本间预训练损失的泛化情况;其次,我们用各嵌入模型引导标准的基于多样性的数据构建算法,通过在所选数据上预训练语言模型并评估下游任务表现来衡量其效用;最后,评估嵌入区分不同数据源样本的能力。\n通过这些评估我们发现,标准现成嵌入模型并不适配预训练数据构建场景,其表现甚至不如从在相同预训练语料上训练的模型中提取的极简嵌入。我们的实验在Pile数据集上开展,用200B token预训练一个1.7B参数的语言模型。我们认为本研究的分析与评估框架为未来设计专门用于推理预训练数据集相似性的嵌入奠定了基础。

英文摘要:

Measuring similarity between training examples is critical for curating high-quality and diverse pretraining datasets for language models. However, similarity is typically computed with a generic off-the-shelf embedding model that has been trained for tasks such as retrieval. Whether these embedding-based similarity metrics are well-suited for pretraining data selection remains largely unexplored. In this paper, we propose a new framework to assess the suitability of a similarity metric specifically for data curation in language model pretraining applications. Our framework's first evaluation criterion captures how well distances reflect generalization in pretraining loss between different training examples. Next, we use each embedding model to guide a standard diversity-based data curation algorithm and measure its utility by pretraining a language model on the selected data and evaluating downstream task performance. Finally, we evaluate the capabilities of embeddings to distinguish between examples from different data sources. With these evaluations, we demonstrate that standard off-the-shelf embedding models are not well-suited for the pretraining data curation setting, underperforming even remarkably simple embeddings that are extracted from models trained on the same pretraining corpus. Our experiments are performed on the Pile, for pretraining a 1.7B parameter language model on 200B tokens. We believe our analysis and evaluation framework serves as a foundation for the future design of embeddings that specifically reason about similarity in pretraining datasets.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑