Beam Search with Bidirectional Strategies for Neural Response Generation
This is an incremental improvement for natural language processing applications like chatbots or dialogue systems.
The paper tackles the problem of suboptimal performance in neural response generation by proposing bidirectional beam search strategies that combine left-to-right and right-to-left language models, demonstrating better performance compared to unidirectional beam search.
Sequence-to-sequence neural networks have been widely used in language-based applications as they have flexible capabilities to learn various language models. However, when seeking for the optimal language response through trained neural networks, current existing approaches such as beam-search decoder strategies are still not able reaching to promising performances. Instead of developing various decoder strategies based on a "regular sentence order" neural network (a trained model by outputting sentences from left-to-right order), we leveraged "reverse" order as additional language model (a trained model by outputting sentences from right-to-left order) which can provide different perspectives for the path finding problems. In this paper, we propose bidirectional strategies in searching paths by combining two networks (left-to-right and right-to-left language models) making a bidirectional beam search possible. Besides, our solution allows us using any similarity measure in our sentence selection criterion. Our approaches demonstrate better performance compared to the unidirectional beam search strategy.