English

Dataset and Neural Recurrent Sequence Labeling Model for Open-Domain Factoid Question Answering

Computation and Language 2016-09-02 v2 Artificial Intelligence Neural and Evolutionary Computing

Abstract

While question answering (QA) with neural network, i.e. neural QA, has achieved promising results in recent years, lacking of large scale real-word QA dataset is still a challenge for developing and evaluating neural QA system. To alleviate this problem, we propose a large scale human annotated real-world QA dataset WebQA with more than 42k questions and 556k evidences. As existing neural QA methods resolve QA either as sequence generation or classification/ranking problem, they face challenges of expensive softmax computation, unseen answers handling or separate candidate answer generation component. In this work, we cast neural QA as a sequence labeling problem and propose an end-to-end sequence labeling model, which overcomes all the above challenges. Experimental results on WebQA show that our model outperforms the baselines significantly with an F1 score of 74.69% with word-based input, and the performance drops only 3.72 F1 points with more challenging character-based input.

Keywords

Cite

@article{arxiv.1607.06275,
  title  = {Dataset and Neural Recurrent Sequence Labeling Model for Open-Domain Factoid Question Answering},
  author = {Peng Li and Wei Li and Zhengyan He and Xuguang Wang and Ying Cao and Jie Zhou and Wei Xu},
  journal= {arXiv preprint arXiv:1607.06275},
  year   = {2016}
}

Comments

10 pages, 3 figures, withdraw experimental results on CNN/Daily Mail datasets

R2 v1 2026-06-22T15:00:24.676Z