arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2012.07280cs.CL
收藏

基于对抗扰动的对比学习条件文本生成

Contrastive Learning with Adversarial Perturbations for Conditional Text Generation

KAIST(韩国科学技术院);AITRICS

Seanie Lee, Dong Bok Lee, Sung Ju Hwang

摘要:

最近,序列到序列(seq 2seq)模型与Transformer架构取得了显着的性能在各种条件文本生成任务,如机器翻译。然而,它们中的大多数都是用教师强迫训练的,在每个时间步都给出了地面真值标签,而不会在训练过程中暴露于错误生成的令牌,这会损害其对看不见的输入的泛化,即所谓的“暴露偏差”问题。在这项工作中,我们建议通过对比正对与负对来减轻条件文本生成问题,使得模型暴露于输入的各种有效或不正确的扰动,以提高泛化能力。然而,使用随机非目标序列作为负例,使用朴素对比学习框架训练模型是次优的,因为它们很容易与正确的输出区分开,特别是使用大型文本语料库预训练的模型。此外,生成正面示例需要特定于域的增强算法,这可能不会在不同的域上推广。为了解决这个问题,我们提出了一个原则性的方法来产生积极的和消极的样本的对比学习的seq 2seq模型。具体来说,我们通过向输入序列添加小扰动来最小化其条件似然来生成负例,并通过添加大扰动来生成正例,同时强制其具有高条件似然。使用我们的方法生成的这种“硬”正负对指导模型更好地区分正确的输出和不正确的输出。我们的经验表明,我们提出的方法显着提高了概括的seq 2seq的三个文本生成任务-机器翻译,文本摘要和问题生成。

英文摘要:

Recently, sequence-to-sequence (seq2seq) models with the Transformer architecture have achieved remarkable performance on various conditional text generation tasks, such as machine translation. However, most of them are trained with teacher forcing with the ground truth label given at each time step, without being exposed to incorrectly generated tokens during training, which hurts its generalization to unseen inputs, that is known as the "exposure bias" problem. In this work, we propose to mitigate the conditional text generation problem by contrasting positive pairs with negative pairs, such that the model is exposed to various valid or incorrect perturbations of the inputs, for improved generalization. However, training the model with naive contrastive learning framework using random non-target sequences as negative examples is suboptimal, since they are easily distinguishable from the correct output, especially so with models pretrained with large text corpora. Also, generating positive examples requires domain-specific augmentation heuristics which may not generalize over diverse domains. To tackle this problem, we propose a principled method to generate positive and negative samples for contrastive learning of seq2seq models. Specifically, we generate negative examples by adding small perturbations to the input sequence to minimize its conditional likelihood, and positive examples by adding large perturbations while enforcing it to have a high conditional likelihood. Such "hard" positive and negative pairs generated using our method guides the model to better distinguish correct outputs from incorrect ones. We empirically show that our proposed method significantly improves the generalization of the seq2seq on three text generation tasks - machine translation, text summarization, and question generation.

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。

↑