arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2206.08446cs.CLcs.AI
收藏

语言模型鲁棒性的评估和改进方法

Methods for Estimating and Improving Robustness of Language Models

Masaryk University(马萨里克大学)

Michal Štefánik

摘要:

尽管大语言模型(LLM)表现出色,但它们存在着臭名昭著的缺陷,即它们偏爱简单的表层文本关系,而不是问题的全部语义复杂性。这个建议调查了这个问题的一个共同点,即他们在训练域之外泛化的能力很弱。我们调查了不同的研究方向,提供模型泛化能力的估计,并发现将这些措施中的一些训练目标,导致增强神经模型的分布鲁棒性。基于这些发现,我们提出了未来的研究方向,以提高鲁棒性的LLM。

英文摘要:

Despite their outstanding performance, large language models (LLMs) suffer notorious flaws related to their preference for simple, surface-level textual relations over full semantic complexity of the problem. This proposal investigates a common denominator of this problem in their weak ability to generalise outside of the training domain. We survey diverse research directions providing estimations of model generalisation ability and find that incorporating some of these measures in the training objectives leads to enhanced distributional robustness of neural models. Based on these findings, we present future research directions towards enhancing the robustness of LLMs.

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。

↑