Corpus ID: 210714186

Modality-Balanced Models for Visual Dialogue

@article{Kim2020ModalityBalancedMF,
  title={Modality-Balanced Models for Visual Dialogue},
  author={Hyounghun Kim and Hao Tan and Mohit Bansal},
  journal={ArXiv},
  year={2020},
  volume={abs/2001.06354}
}
  • Hyounghun Kim, Hao Tan, Mohit Bansal
  • Published in AAAI 2020
  • Computer Science
  • ArXiv
  • The Visual Dialog task requires a model to exploit both image and conversational context information to generate the next response to the dialogue. However, via manual analysis, we find that a large number of conversational questions can be answered by only looking at the image without any access to the context history, while others still need the conversation context to predict the correct answers. We demonstrate that due to this reason, previous joint-modality (history and image) models over… CONTINUE READING

    Create an AI-powered research feed to stay up to date with new papers like this posted to ArXiv

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 28 REFERENCES

    J

    • A. Das, S. Kottur, +3 authors Moura
    • M.; Parikh, D.; and Batra, D.
    • 2017
    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    R

    • B. Efron, Tibshirani
    • J.
    • 1994
    VIEW 1 EXCERPT
    HIGHLY INFLUENTIAL

    A

    • Z. Gan, Y. Cheng, Kholy
    • E.; Li, L.; Liu, J.; and Gao, J.
    • 2019
    VIEW 2 EXCERPTS

    Image-Question-Answer Synergistic Network for Visual Dialog

    Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering

    J

    • S. Kottur, Moura
    • M.; Parikh, D.; Batra, D.; and Rohrbach, M.
    • 2018

    P

    • D. Massiceti, P. K. Dokania, N. Siddharth, Torr
    • H.
    • 2018
    VIEW 1 EXCERPT

    Recursive Visual Attention in Visual Dialog