A non-hierarchical attention network with modality dropout for textual response generation in multimodal dialogue systems
This work improves multimodal dialogue systems for applications like virtual assistants by offering a novel method to enhance response generation, though it is incremental as it builds on prior attention-based approaches.
The paper tackled the problem of generating textual responses in multimodal dialogue systems by addressing insufficient fine-grained interaction between textual and visual features and incomplete context representation in existing hierarchical frameworks, proposing a non-hierarchical attention network with modality dropout that achieved state-of-the-art performance on a public dataset.
Existing text- and image-based multimodal dialogue systems use the traditional Hierarchical Recurrent Encoder-Decoder (HRED) framework, which has an utterance-level encoder to model utterance representation and a context-level encoder to model context representation. Although pioneer efforts have shown promising performances, they still suffer from the following challenges: (1) the interaction between textual features and visual features is not fine-grained enough. (2) the context representation can not provide a complete representation for the context. To address the issues mentioned above, we propose a non-hierarchical attention network with modality dropout, which abandons the HRED framework and utilizes attention modules to encode each utterance and model the context representation. To evaluate our proposed model, we conduct comprehensive experiments on a public multimodal dialogue dataset. Automatic and human evaluation demonstrate that our proposed model outperforms the existing methods and achieves state-of-the-art performance.