Dynamic Coattention Networks For Question Answering

Dynamic Coattention Networks For Question Answering

Caiming Xiong,  Victor Zhong,  Richard Socher

(Submitted on 5 Nov 2016 ( v1), last revised 13 Feb 2017 (this version, v3))

Several deep learning models have been proposed for question answering. However, due to their single-pass nature, they have no way to recover from local maxima corresponding to incorrect answers. To address this problem, we introduce the Dynamic Coattention Network (DCN) for question answering. The DCN first fuses co-dependent representations of the question and the document in order to focus on relevant parts of both. Then a dynamic pointing decoder iterates over potential answer spans. This iterative procedure enables the model to recover from initial local maxima corresponding to incorrect answers. On the Stanford question answering dataset, a single DCN model improves the previous state of the art from 71.0% F1 to 75.9%, while a DCN ensemble obtains 80.4% F1.

Comments:	14 pages, 7 figures, International Conference on Learning Representations 2017
Subjects:	Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
Cite as:	arXiv:1611.01604 [cs.CL]
	(or arXiv:1611.01604v3 [cs.CL] for this version)

Submission history

From: Victor Zhong [ view email] 
[v1] Sat, 5 Nov 2016 04:53:40 GMT (802kb,D)
[v2] Thu, 17 Nov 2016 19:58:22 GMT (803kb,D)
[v3] Mon, 13 Feb 2017 23:00:32 GMT (803kb,D)