arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2311.04166cs.CLcs.LG
收藏

扰动样本揭示语言模型共享的不变性

Perturbed examples reveal invariances shared by language models

MPI for Software Systems(马克斯·普朗克软件系统研究所)

Ruchit Rawal, Mariya Toneva

AI总结:

提出一种通过可解释输入扰动比较两个NLP模型共享不变性的框架,用于揭示蒸馏、规模变化及黑盒模型间语言能力的异同。

摘要:

自然语言处理(NLP)研究的快速增长催生了大量新模型,其速度超过了我们理解它们与既有模型之间差异的能力。造成这一困难的一个主要原因是基准测试趋于饱和,可能无法很好地反映模型在真实环境中的性能差异。在本研究中,我们提出了一个新颖框架,通过揭示两个NLP模型对针对特定语言能力的可解释输入扰动所共有的不变性,来比较它们。通过对来自相同和不同架构家族的模型进行实验,该框架为模型变化(例如蒸馏、规模增大)如何影响语言能力提供了洞见。此外,我们的框架还能够评估商业黑盒模型(例如InstructGPT家族)与被更充分理解的模型(例如GPT-2)之间的不变性。在所有实验中,我们观察到大型语言模型共享由各种规模模型编码的许多不变性,而大型模型具有的不变性仅被其他大型模型共享。拥有多种多样的不变性可能是大型语言模型近期成功的关键,并且我们的框架能够阐明新模型中保留或涌现的不变性类型。我们公开发布了代码。

英文摘要:

The rapid growth in natural language processing (NLP) research has led to numerous new models, outpacing our understanding of how they compare to established ones. One major reason for this difficulty is saturating benchmarks, which may not well reflect differences in model performance in the wild. In this work, we introduce a novel framework to compare two NLP models by revealing their shared invariance to interpretable input perturbations targeting a specific linguistic capability. Via experiments on models from the same and different architecture families, this framework offers insights about how changes in models (e.g., distillation, size increase) affect linguistic capabilities. Furthermore, our framework enables evaluation of invariances between commercial black-box models (e.g., InstructGPT family) and models that are better understood (e.g., GPT-2). Across experiments, we observe that large language models share many invariances encoded by models of various sizes, whereas the invariances by large models are only shared by other large models. Possessing a wide variety of invariances may be key to the recent successes of large language models, and our framework can shed light on the types of invariances retained or emerging in new models. We make the code publicly available.

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。

↑