Towards Less Generic Responses in Neural Conversation Models: A Statistical Re-weighting Method
Yahui Liu, Wei Bi, Jun Gao, Xiaojiang Liu, Jian Gang Yao, Shuming Shi · 2018
Sequence-to-sequence neural generation models have achieved promising performance on short text conversation tasks.However, they tend to generate generic/dull responses, leading to unsatisfying dialogue experience.We observe that in conversation tasks, each query could have multiple responses, which forms a 1-to-n or m-to-n relationship in the view of the total corpus.The objective function used in standard sequence-to-sequence models will be dominated by loss terms with generic patterns.Inspired by this observation, we introduce a statistical re-weighting method that assigns different weights for the multiple responses of the same query, and trains the standard neural generation model with the weights.Experimental results on a large Chinese dialogue corpus show that our method improves the acceptance rate of generated responses compared with several baseline models and significantly reduces the number of generated generic responses.