Learning to Rank Question-Answer Pairs using Hierarchical Recurrent Encoder with Latent Topic Clustering

NAACL 2018 Seunghyun YoonJoongbo ShinKyomin Jung

In this paper, we propose a novel end-to-end neural architecture for ranking candidate answers, that adapts a hierarchical recurrent neural network and a latent topic clustering module. With our proposed model, a text is encoded to a vector representation from an word-level to a chunk-level to effectively capture the entire meaning... (read more)

PDF Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT LEADERBOARD
Answer Selection Ubuntu Dialogue (v1, Ranking) HRDE-LTC 1 in 10 [email protected] 0.684 # 1
1 in 2 [email protected] 0.916 # 1
1 in 10 [email protected] 0.822 # 1
1 in 10 [email protected] 0.960 # 1
Answer Selection Ubuntu Dialogue (v2, Ranking) HRDE-LTC 1 in 10 [email protected] 0.652 # 1
1 in 10 [email protected] 0.815 # 1
1 in 10 [email protected] 0.966 # 1
1 in 2 [email protected] 0.915 # 1