A Simple, Fast Diverse Decoding Algorithm for Neural Generation

Abstract

In this paper, we propose a simple, fast decoding algorithm that fosters diversity in neural generation. The algorithm modifies the standard beam search algorithm by adding an inter-sibling ranking penalty, favoring choosing hypotheses from diverse parents. We evaluate the proposed model on the tasks of dialogue response generation, abstractive summarization and machine translation. We find that diverse decoding helps across all tasks, especially those for which reranking is needed. We further propose a variation that is capable of automatically adjusting its diversity decoding rates for different inputs using reinforcement learning (RL). We observe a further performance boost from this RL technique. This paper includes material from the unpublished script "Mutual Information and Diverse Decoding Improve Neural Machine Translation" (Li and Jurafsky, 2016).

References in corpus (6)

• Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
• Microsoft COCO Captions: Data Collection and Evaluation Server
• On Using Monolingual Corpora in Neural Machine Translation
• A Hierarchical Neural Autoencoder for Paragraphs and Documents
• A Neural Network Approach to Context-Sensitive Generation of Conversational Responses
• Learning to Compose Words into Sentences with Reinforcement Learning

Cited by in corpus (6)

• Learning to Decode for Future Success
• Neural Text Generation: A Practical Guide
• Trainable Greedy Decoding for Neural Machine Translation
• Generating Multiple Diverse Responses with Multi-Mapping and Posterior Mapping Selection
• Affective Neural Response Generation
• Single-Queue Decoding for Neural Machine Translation