An animated picture says at least a thousand words: Selecting Gif-based Replies in Multimodal Dialog.
Xingyao WangDavid JurgensPublished in: EMNLP (Findings) (2021)
Keyphrases
- multi modal
- mixed initiative
- keywords
- natural language
- multimodal information
- user interface
- text documents
- word sense disambiguation
- conversational agents
- syntactic categories
- selection algorithm
- word recognition
- tens of thousands
- audio visual
- word meaning
- stop words
- text corpora
- multi party
- database
- n gram
- multimedia