The production of color language is essential for grounded language generation. Color descriptions have many challenging properties: they can be vague, compositionally complex, and denotationally rich. We present an effective approach to generating color descriptions using recurrent neural networks and a Fourier-transformed color representation. Our model outperforms previous work on a conditional language modeling task over a large corpus of naturalistic color descriptions. In addition, probing the model's output reveals that it can accurately produce not only basic color terms but also descriptors with non-convex denotations ("greenish"), bare modifiers ("bright", "dull"), and compositional phrases ("faded teal") not seen in training.
@article{arxiv.1606.03821,
title = {Learning to Generate Compositional Color Descriptions},
author = {Will Monroe and Noah D. Goodman and Christopher Potts},
journal= {arXiv preprint arXiv:1606.03821},
year = {2016}
}