Natural Language Inference by Tree-Based Convolution and Heuristic Matching
This addresses natural language inference for NLP researchers, but it is incremental as it builds on existing tree-based and matching techniques.
The paper tackled the problem of recognizing entailment and contradiction between two sentences by proposing the TBCNN-pair model, which uses a tree-based convolutional neural network and heuristic matching layers, and it outperformed existing sentence encoding-based approaches by a large margin.
In this paper, we propose the TBCNN-pair model to recognize entailment and contradiction between two sentences. In our model, a tree-based convolutional neural network (TBCNN) captures sentence-level semantics; then heuristic matching layers like concatenation, element-wise product/difference combine the information in individual sentences. Experimental results show that our model outperforms existing sentence encoding-based approaches by a large margin.