Learning to Generate Compositional Color Descriptions

@inproceedings{Monroe2016LearningTG,
  title={Learning to Generate Compositional Color Descriptions},
  author={Will Monroe and Noah D. Goodman and Christopher Potts},
  booktitle={EMNLP},
  year={2016}
}
  • Will Monroe, Noah D. Goodman, Christopher Potts
  • Published in EMNLP 2016
  • Computer Science
  • 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… CONTINUE READING

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 17 CITATIONS

    Grounded Learning of Color Semantics with Autoencoders

    VIEW 9 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    Learning to Compose Spatial Relations with Grounded Neural Language Models

    VIEW 2 EXCERPTS
    CITES METHODS
    HIGHLY INFLUENCED

    Colors in Context: A Pragmatic Neural Model for Grounded Language Understanding

    VIEW 1 EXCERPT
    CITES METHODS

    Learning Distributions of Meant Color

    VIEW 7 EXCERPTS
    CITES BACKGROUND & RESULTS
    HIGHLY INFLUENCED

    Language-based colorization of scene sketches

    VIEW 1 EXCERPT
    CITES RESULTS

    Compositionality for perceptual classification

    VIEW 3 EXCERPTS
    CITES BACKGROUND

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 25 REFERENCES