arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.08124cs.CL
收藏

XFBoost:使用可控解码器改进文本生成

XFBoost: Improving Text Generation with Controllable Decoders

Georgia Institute of Technology(佐治亚理工学院);Salesforce Einstein(Salesforce爱因斯坦)

Xiangyu Peng, Michael Sollami

摘要:

基于transformer的自然语言模型中的多模态条件性在产品描述生成任务中表现出了最先进的性能。最近的方法使语言模型以一个或多个图像和其他文本元数据为条件,以实现用于描述来自电子商务商店的产品的接近人类的性能。然而,生成的描述可能会表现出一定程度的不准确性,甚至相对于给定产品的输入相互矛盾的声明。在本文中,我们提出了一个可控的语言生成框架,称为提取Finetune Boost(XFBoost),它解决了不准确的低质量推理的问题。通过在生成过程的解码阶段使用视觉语义属性作为约束条件,并使用策略梯度技术微调语言模型,发现XFBoost框架可以生成具有更高图像相关性的描述性文本,优于基线,并降低事实上不准确描述的频率。我们进一步展示了XFBoost在在线学习中的应用,其中人在回路中的批评者通过主动反馈改进语言模型。

英文摘要:

Multimodal conditionality in transformer-based natural language models has demonstrated state-of-the-art performance in the task of product description generation. Recent approaches condition a language model on one or more images and other textual metadata to achieve near-human performance for describing products from e-commerce stores. However, generated descriptions may exhibit degrees of inaccuracy or even contradictory claims relative to the inputs of a given product. In this paper, we propose a controllable language generation framework called Extract-Finetune-Boost (XFBoost), which addresses the problem of inaccurate low-quality inference. By using visual semantic attributes as constraints at the decoding stage of the generation process and finetuning the language model with policy gradient techniques, the XFBoost framework is found to produce significantly more descriptive text with higher image relevancy, outperforming baselines and lowering the frequency of factually inaccurate descriptions. We further demonstrate the application of XFBoost to online learning wherein human-in-the-loop critics improve language models with active feedback.

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。

↑