Are Current Decoding Strategies Capable of Facing the Challenges of Visual Dialogue?

Open in new window