Multimedia dialogue systems have become more prevalent in various industries, including travel, retail, and others. Multimedia systems have gotten less attention than text media, although most previous research on conversational systems has exclusively concentrated on text. In addition to obtaining outstanding results in earlier studies, other difficulties—such as selecting the appropriate medium for the response and retrieving the most suitable image throughout the conversation—have received less attention. This thesis used the high-end image question answering approaches to address these issues, significantly improving the results of earlier multimedia dialogue system models in terms of the image matching criterion.
-
Notifications
You must be signed in to change notification settings - Fork 1
Multi Modal Task Oriented Dialogue System (MMTOD)
License
ahkarimi/MMTOD
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
About
Multi Modal Task Oriented Dialogue System (MMTOD)