Towards Less Generic Responses in Neural Conversation Models: A Statistical Re-weighting Method.
Yahui LiuWei BiJun GaoXiaojiang LiuJian YaoShuming ShiPublished in: EMNLP (2018)
Keyphrases
- probabilistic model
- statistical methods
- mutual information
- computational cost
- weighting scheme
- monte carlo simulation
- segmentation method
- experimental evaluation
- pairwise
- computational complexity
- prior knowledge
- statistical information
- dynamic programming
- high accuracy
- neural network
- high precision
- statistical model
- generic model
- statistical approaches
- statistical measures
- clustering method
- parameter estimation
- significant improvement
- k means
- preprocessing
- bayesian networks
- similarity measure
- genetic algorithm
- machine learning