arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2008.09075cs.CLcs.AI
收藏

用语义范例控制对话生成

Controlling Dialogue Generation with Semantic Exemplars

Carnegie Mellon University(卡内基梅隆大学)

Prakhar Gupta, Jeffrey P. Bigham, Yulia Tsvetkov, Amy Pavel

摘要:

用大语言模型预训练的对话系统生成局部连贯的响应,但缺乏对实现特定目标所需的响应的细粒度控制。一个有前途的方法来控制响应生成是基于范例的生成,其中模型编辑范例响应,从训练数据检索,或手写的战略地址话语级的目标,以适应新的对话上下文。但是,目前基于范例的方法往往过度复制的话从范例的反应,导致不连贯的答复。我们提出了一个基于范例的对话生成模型,EDGE,它使用的语义框架中存在的范例响应,以指导生成。我们表明,控制对话生成的基础上的语义框架的范例,而不是在范例本身的话,提高了生成的响应的一致性,同时保持语义意义和会话目标存在于范例响应。

英文摘要:

Dialogue systems pretrained with large language models generate locally coherent responses, but lack the fine-grained control over responses necessary to achieve specific goals. A promising method to control response generation is exemplar-based generation, in which models edit exemplar responses that are retrieved from training data, or hand-written to strategically address discourse-level goals, to fit new dialogue contexts. But, current exemplar-based approaches often excessively copy words from the exemplar responses, leading to incoherent replies. We present an Exemplar-based Dialogue Generation model, EDGE, that uses the semantic frames present in exemplar responses to guide generation. We show that controlling dialogue generation based on the semantic frames of exemplars, rather than words in the exemplar itself, improves the coherence of generated responses, while preserving semantic meaning and conversation goals present in exemplar responses.

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。

↑