arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2107.00440cs.CLcs.AI
收藏

CLINE:基于语义否定实例的对比学习及其在自然语言理解中的应用

CLINE: Contrastive Learning with Semantic Negative Examples for Natural Language Understanding

Tsinghua University(清华大学);Tsinghua ShenZhen International Graduate School, Tsinghua University(清华大学深圳国际研究生院);Tencent AI Lab(腾讯AI Lab)

Dong Wang, Ning Ding, Piji Li, Hai-Tao Zheng

摘要:

尽管预先训练的语言模型已被证明对学习高质量的语义表示有用,但这些模型仍然容易受到简单扰动的影响。最近旨在提高预训练模型鲁棒性的工作主要集中在从具有相似语义的扰动样本中进行对抗性训练,忽略了不同甚至相反语义的利用。与图像处理领域不同,文本是离散的,很少的单词替换就会引起显著的语义变化。为了研究小扰动对语义的影响,我们进行了一系列的试点实验,令人惊讶地发现对抗性训练对于模型检测这些语义变化是无用的,甚至是有害的。针对这一问题,本文提出了基于语义负例的对比学习(CLINE)方法,该方法在无监督的情况下构造语义负例,以提高算法在语义对抗攻击下的鲁棒性。通过与语义相似和相反的例子进行比较,该模型可以有效地感知小扰动引起的语义变化。实证结果表明,我们的方法产生了一系列的情感分析,推理和阅读理解任务的实质性改善。CLINE还保证了同一语义内的紧凑性和不同语义间的可分离性。

英文摘要:

Despite pre-trained language models have proven useful for learning high-quality semantic representations, these models are still vulnerable to simple perturbations. Recent works aimed to improve the robustness of pre-trained models mainly focus on adversarial training from perturbed examples with similar semantics, neglecting the utilization of different or even opposite semantics. Different from the image processing field, the text is discrete and few word substitutions can cause significant semantic changes. To study the impact of semantics caused by small perturbations, we conduct a series of pilot experiments and surprisingly find that adversarial training is useless or even harmful for the model to detect these semantic changes. To address this problem, we propose Contrastive Learning with semantIc Negative Examples (CLINE), which constructs semantic negative examples unsupervised to improve the robustness under semantically adversarial attacking. By comparing with similar and opposite semantic examples, the model can effectively perceive the semantic changes caused by small perturbations. Empirical results show that our approach yields substantial improvements on a range of sentiment analysis, reasoning, and reading comprehension tasks. And CLINE also ensures the compactness within the same semantics and separability across different semantics in sentence-level.

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。

↑