# Neural-based Natural Language Generation in Dialogue using RNN Encoder-Decoder with Semantic Aggregation

## TL;DR

This paper introduces an Encoder-Aggregator-Decoder neural model with semantic aggregation for natural language generation in dialogue systems, demonstrating improved performance across multiple domains.

## Contribution

It proposes a novel Semantic Aggregator with attention and gating mechanisms, enhancing sentence planning and surface realization in neural NLG models.

## Key findings

- Consistently outperforms previous methods on four NLG domains
- Jointly trains sentence planning and surface realization
- Effective semantic aggregation improves natural language output

## Abstract

Natural language generation (NLG) is an important component in spoken dialogue systems. This paper presents a model called Encoder-Aggregator-Decoder which is an extension of an Recurrent Neural Network based Encoder-Decoder architecture. The proposed Semantic Aggregator consists of two components: an Aligner and a Refiner. The Aligner is a conventional attention calculated over the encoded input information, while the Refiner is another attention or gating mechanism stacked over the attentive Aligner in order to further select and aggregate the semantic elements. The proposed model can be jointly trained both sentence planning and surface realization to produce natural language utterances. The model was extensively assessed on four different NLG domains, in which the experimental results showed that the proposed generator consistently outperforms the previous methods on all the NLG domains.