Reasoning about Pragmatics with Neural Listeners and Speakers
Reasoning about Pragmatics with Neural Listeners and Speakers
复制标题
DOI：
10.18653/v1/d16-1125
复制
发表时间：
2016-04
期刊：
影响因子：
--
通讯作者：
Jacob Andreas;D. Klein
中科院分区：
文献类型：
--
作者：
Jacob Andreas;D. Klein
We present a model for pragmatically describing scenes, in which contrastive behavior results from a combination of inference-driven pragmatics and learned semantics. Like previous learned approaches to language generation, our model uses a simple feature-driven architecture (here a pair of neural "listener" and "speaker" models) to ground language in the world. Like inference-driven approaches to pragmatics, our model actively reasons about listener behavior when selecting utterances. For training, our approach requires only ordinary captions, annotated _without_ demonstration of the pragmatic behavior the model ultimately exhibits. In human evaluations on a referring expression game, our approach succeeds 81% of the time, compared to a 69% success rate using existing techniques.
