Review papers:
A Joint Speaker-Listener-Reinforcer Model for Referring Expressions:
https://arxiv.org/pdf/1612.09542v1.pdf
Context-aware Captions from Context-agnostic Supervision:
https://arxiv.org/pdf/1701.02870.pdf
Extra papers:
An Actor-Critic Algorithm for Sequence Prediction:
https://arxiv.org/pdf/1607.07086v2.pdf
Sequence-to-Sequence Learning as Beam-Search Optimization:
https://arxiv.org/abs/1606.02960
Reasoning about Pragmatics with Neural Listeners and Speakers: