Abstract:Natural language sentence matching is the task of comparing two sentences and identifying the relationship between them. It has a wide range of applications in natural language processing tasks such as reading comprehension, question and answer systems. The main approach is to compute the interaction between text representations and sentence pairs through an attention mechanism, which can extract the semantic information between sentence pairs well. However, this kind of methods fail to capture deep semantic information and effectively fuse the semantic information of the sentence. To solve this problem, we propose a sentence matching method based on deep interaction and fusion. We first use pre-trained word vectors Glove and character-level word vectors to obtain word embedding representations of the two sentences. In the encoding layer, we use bidirectional LSTM to encode the sentence pairs. In the interaction layer, we initially fuse the information of the sentence pairs to obtain low-level semantic information; at the same time, we use the bi-directional attention in the machine reading comprehension model and self-attention to obtain the high-level semantic information. We use a heuristic fusion function to fuse the low-level semantic information and the high-level semantic information to obtain the final semantic information, and finally we use the neural network to predict the answer. We evaluate our model on two tasks: text implication recognition and paraphrase recognition. We conducted experiments on the SNLI datasets for the recognizing textual entailment task, the Quora dataset for the paraphrase recognition task. The experimental results show that the proposed algorithm can effectively fuse different semantic information that verify the effectiveness of the algorithm on sentence matching tasks.

Deep and Shallow Features Learning for Short Texts Matching

A Short Texts Matching Method Using Shallow Features and Deep Features

A Survey on Deep Text Matching

Syntax-based Deep Matching of Short Texts

Deep Learning for Text Matching: A Survey

Lightweight Text Matching Method with Rich Features.

Cross-lingual Short-text Matching with Deep Learning

Attention-Based Multi-level Network for Text Matching with Feature Fusion

RI-Match: Integrating Both Representations and Interactions for Deep Semantic Matching.

Deep Fusion LSTMs for Text Semantic Matching.

MatchZoo: A Toolkit for Deep Text Matching.

Short text matching model with multiway semantic interaction based on multi-granularity semantic embedding

A simple and efficient text matching model based on deep interaction

Sentence Matching with Deep Self-attention and Co-attention Features

Advanced Multimodal Deep Learning Architecture for Image-Text Matching

DIFM: an Effective Deep Interaction and Fusion Model for Sentence Matching

Neural Graph Matching Networks for Chinese Short Text Matching.

Simple and Effective Text Matching with Richer Alignment Features

The Short Text Matching Model Enhanced with Knowledge via Contrastive Learning

Deep Feature Matching for Dense Correspondence