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 2406.14005cs.CLcs.AIcs.LG
收藏

面向语言模型微调的信息引导正则化

Information Guided Regularization for Fine-tuning Language Models

Virginia Tech(弗吉尼亚理工大学);Stevens Institute of Technology(史蒂文斯理工学院)

Mandar Sharma, Nikhil Muralidhar, Shengzhe Xu, Raquib Bin Yousuf, Naren Ramakrishnan

AI总结:

该研究针对语言模型迁移学习的正则化需求,从信息论视角分析任务敏感参数对预训练损失景观的影响,提出任务与架构无关、无额外计算开销的guided dropout方法,在数据稀缺等场景下相比标准基线取得更优的下游性能。

摘要:

预训练-微调范式已成为现代语言建模中迁移学习的事实上的标准策略。基于语言模型中的任务适配通常是跨任务共享参数的函数这一认知,我们认为需要一种更精准的正则化方法以实现更平滑的迁移学习。为此,我们从信息论视角研究了预训练损失景观如何受这些任务敏感参数的影响。随后,我们利用研究发现设计了一种新颖的dropout方法,以改进模型正则化并提升下游泛化能力。该方法名为guided dropout(引导式dropout),兼具任务与架构无关性,且不会给微调过程增加计算开销。通过实证评估,我们证明与标准基线相比,我们的正则化方法即使在数据稀缺场景下也能带来持续更优的性能。

英文摘要:

The pretraining-fine-tuning paradigm has been the de facto strategy for transfer learning in modern language modeling. With the understanding that task adaptation in LMs is often a function of parameters shared across tasks, we argue that a more surgical approach to regularization needs to exist for smoother transfer learning. Towards this end, we investigate how the pretraining loss landscape is affected by these task-sensitive parameters through an information-theoretic lens. We then leverage the findings from our investigations to devise a novel approach to dropout for improved model regularization and better downstream generalization. This approach, named guided dropout, is both task & architecture agnostic and adds no computational overhead to the fine-tuning process. Through empirical evaluations, we showcase that our approach to regularization yields consistently better performance, even in scenarios of data paucity, compared to standardized baselines.

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。

↑