On Tree-Based Neural Sentence Modeling

EMNLP 2018  ·  Haoyue Shi, Hao Zhou, Jiaze Chen, Lei LI ·

Neural networks with tree-based sentence encoders have shown better results on many downstream tasks. Most of existing tree-based encoders adopt syntactic parsing trees as the explicit structure prior. To study the effectiveness of different tree structures, we replace the parsing trees with trivial trees (i.e., binary balanced tree, left-branching tree and right-branching tree) in the encoders. Though trivial trees contain no syntactic information, those encoders get competitive or even better results on all of the ten downstream tasks we investigated. This surprising result indicates that explicit syntax guidance may not be the main contributor to the superior performances of tree-based neural sentence modeling. Further analysis show that tree modeling gives better results when crucial words are closer to the final representation. Additional experiments give more clues on how to design an effective tree-based encoder. Our code is open-source and available at https://github.com/ExplorerFreda/TreeEnc.

PDF Abstract EMNLP 2018 PDF EMNLP 2018 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Text Classification AG News Balanced+bi-leaf-RNN Error 7.9 # 16
Sentiment Analysis Amazon Review Full Gumbel+bi-leaf-RNN Accuracy 49.7 # 9
Sentiment Analysis Amazon Review Polarity Gumbel+bi-leaf-RNN Accuracy 88.1 # 9
Text Classification DBpedia Balanced+bi-leaf-RNN Error 1.2 # 15

Methods


No methods listed for this paper. Add relevant methods here