Affective Neural Response Generation

Existing neural conversational models process natural language primarily on a lexico-syntactic level, thereby ignoring one of the most crucial components of human-to-human dialogue: its affective content. We take a step in this direction by proposing three novel ways to incorporate affective/emotional aspects into long short term memory (LSTM) encoder-decoder neural conversation models: (1) affective word embeddings, which are cognitively engineered, (2) affect-based objective functions that augment the standard cross-entropy loss, and (3) affectively diverse beam search for decoding. Experiments show that these techniques improve the open-domain conversational prowess of encoderdecoder networks by enabling them to produce more natural and emotionally rich responses.

现有的神经对话模型主要在词汇 - 句法层面处理自然语言，从而忽略了人与人对话中最关键的组成部分之一：其情感内容。我们朝着这个方向迈出了一步，提出了三种将情感方面纳入长短期记忆（LSTM）编码器 - 解码器神经对话模型的新方法：（1）情感词嵌入，这是经过认知设计的；（2）基于情感的目标函数，它增强了标准的交叉熵损失；（3）用于解码的情感多样化束搜索。实验表明，这些技术通过使编码器 - 解码器网络能够生成更自然且情感更丰富的回应，提高了它们在开放领域的对话能力。