Title | ||
---|---|---|
MS-Ranker: Accumulating evidence from potentially correct candidates via reinforcement learning for answer selection |
Abstract | ||
---|---|---|
Answer selection (AS) aims to select correct answers for a question from an answer candidate set. Conventional AS methods generally address this task by independently matching the question and each candidate. However, since the matching information between the question and a single candidate is usually limited, it is not enough to use the question as the only evidence to estimate the correctness of each candidate. To address this problem, we propose a novel reinforcement learning (RL) based multi-step ranking model, named MS-Ranker, which accumulates candidate. In specific, we explicitly consider the potential correctness of candidates when accumulating information and update the evidence with a gating mechanism. Moreover, as we use a listwise ranking reward, our model learns to pay more attention to the overall performance. Experiments on three benchmarks, namely WikiQA, SemEval-2016 CQA and SelQA, show that our model significantly outperforms existing methods that do not rely on external resources. |
Year | DOI | Venue |
---|---|---|
2021 | 10.1016/j.neucom.2021.03.083 | Neurocomputing |
Keywords | DocType | Volume |
Answer selection,Reinforcement learning,MS-Ranker,Gating mechanism,Listwise ranking reward | Journal | 449 |
ISSN | Citations | PageRank |
0925-2312 | 0 | 0.34 |
References | Authors | |
0 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yingxue Zhang | 1 | 0 | 0.68 |
Fandong Meng | 2 | 0 | 0.34 |
Peng Li | 3 | 146 | 21.34 |
Ping Jian | 4 | 4 | 1.77 |
Jie Zhou | 5 | 13 | 11.09 |