# Natural Language Generation in Dialogue using Lexicalized and Delexicalized Data

arXiv: https://arxiv.org/abs/1606.03632 · HTML: https://mlsanity.com/paper/1606.03632/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Jun 2016.

## Built on by (most cited first, 2 of 2)

- [Dual Hierarchical Dialogue Policy Learning for Legal Inquisitive Conversational Agents](https://mlsanity.com/paper/2605.14057/) (2026)
- [Verbal behavior without syntactic structures: beyond Skinner and Chomsky](https://mlsanity.com/paper/2303.08080/) (2023)

## Abstract (arXiv)

Natural language generation plays a critical role in spoken dialogue systems. We present a new approach to natural language generation for task-oriented dialogue using recurrent neural networks in an encoder-decoder framework. In contrast to previous work, our model uses both lexicalized and delexicalized components i.e. slot-value pairs for dialogue acts, with slots and corresponding values aligned together. This allows our model to learn from all available data including the slot-value pairing, rather than being restricted to delexicalized slots. We show that this helps our model generate more natural sentences with better grammar. We further improve our model's performance by transferring weights learnt from a pretrained sentence auto-encoder. Human evaluation of our best-performing model indicates that it generates sentences which users find more appealing.