Neural Paraphrase Identification of Questions with Noisy Pretraining
arXiv:1704.04565
Abstract
We present a solution to the problem of paraphrase identification of questions. We focus on a recent dataset of question pairs annotated with binary paraphrase labels and show that a variant of the decomposable attention model (Parikh et al., 2016) results in accurate performance on this task, while being far simpler than many competing neural architectures. Furthermore, when the model is pretrained on a noisy dataset of automatically collected question paraphrases, it obtains the best reported performance on the dataset.
References in corpus (5)
- Character-Aware Neural Language Models
- A large annotated corpus for learning natural language inference
- Sentence Similarity Learning by Lexical Decomposition and Composition
- Bilateral Multi-Perspective Matching for Natural Language Sentences
- Charagram: Embedding Words and Sentences via Character n-grams
Cited by in corpus (6)
- A Deep Network Model for Paraphrase Detection in Short Text Messages
- Learning General Purpose Distributed Sentence Representations via Large Scale Multi-task Learning
- Stochastic Answer Networks for Natural Language Inference
- Using Prior Knowledge to Guide BERT's Attention in Semantic Textual Matching Tasks
- Semantic Sentence Matching with Densely-connected Recurrent and Co-attentive Information
- Sentence Encoding with Tree-constrained Relation Networks