Bilateral Multi-Perspective Matching for Natural Language Sentences

By Zhiguo Wang, Wael Hamza, Radu Florian. 2017

843 citations recorded by the source corpus.

Natural language sentence matching is a fundamental technology for a variety of tasks. Previous approaches either match sentences from a single direction or only apply single granular (word-by-word or sentence-by-sentence) matching. In this work, we propose a bilateral multi-perspective matching (BiMPM) model under the "matching-aggregation" framework. Given two sentences $P$ and $Q$, our model first encodes them with a BiLSTM encoder. Next, we match the two encoded sentences in two directions $P \rightarrow Q$ and $P \leftarrow Q$. In each matching direction, each time step of one sentence is matched against all time-steps of the other sentence from multiple perspectives. Then, another BiLSTM layer is utilized to aggregate the matching results into a fix-length matching vector. Finally, based on the matching vector, the decision is made through a fully connected layer. We evaluate our model on three tasks: paraphrase identification, natural language inference and answer sentence selection. Experimental results on standard benchmark datasets show that our model achieves the state-of-the-art performance on all tasks.

Subfields: Artificial Intelligence, Natural Language Processing, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Natural Language Processing

In Evaluation & Benchmarks

Citations in the corpus

The Natural Language Decathlon: Multitask Learning as Question Answering (2018, background): “…hold for question answering in general, so recent models for SQuAD are not necessarily general question answering models [Yu et al., 2018a, Hu et al., 2018, Wang et al., 2017a, Liu et al., 2017b, Huang et al., 2017, Xiong et al., 2018, Liu et al., 2017a, Pan et al., 2017, Salant and Berant, 2017].”
Natural Language Inference over Interaction Space (2017, methodology): “N QUORA QUESTION PAIR DATASET In this subsection, we evaluate the effectiveness of our model for paraphrase identiﬁcation as natural language inference task. Other than our baselines, we compare with Wang et al. (2017) and 8 Model Accuracy Dev Acc Test Acc 1. Siamese-CNN - 79.60 2. Multi-Perspective”
Pathologies of Neural Models Make Interpretations Difficult (2018, methodology): “Perspective Matching model (Wang et al., 2017)) from each class of SNLI through the Decomposable Attention Model (Parikh et al.”
Question Answering through Transfer Learning from Large Fine-grained Supervision Data (2017, neutral): “%) by Wang et al. (2017).”
Shortcut-Stacked Sentence Encoders for Multi-Domain Inference (2017, methodology): “…Network (SPINN) in Bowman et al. (2016), and (2) joint, pairwise models that use cross-features be-tween the two sentences to encode them, such as the Enhanced Sequential Inference Model (ESIM) in Chen et al. (2017) or the bilateral multi-perspective matching (BiMPM) model Wang et al. (2017).”
Universal Natural Language Processing with Limited Annotations: Try Few-shot Textual Entailment as a Start (2020, background): “Representative work includes the ﬁrst attentive recurrent neural network (Rockt¨aschel et al., 2016) and its followers (Wang and Jiang, 2016; Wang et al., 2017), as well as the attentive convolutional networks such as attentive pooling (dos Santos et al., 2016) and attentive convolution (Yin and…”
Identifying Well-formed Natural Language Questions (2018, methodology): “…are available tools and datasets, such as Quora question paraphrases and the Par-alex dataset (Fader et al., 2013) – for identifying query paraphrases (Wang et al., 2017; Tomar et al., 2017), but these datasets do not contain information about whether a query is a natural language question or not.”
Challenges in Building Intelligent Open-domain Dialog Systems (2019, background): “Other matching models that were proposed originally for non-dialog tasks such as paraphrase detection, language inference, and reading comprehension [64, 101], have also”
Improved Neural Relation Detection for Knowledge Base Question Answering (2017, background): “We hypothesize that the idea of hierarchical matching with attention mechanism may work better for long sequences, and the new advanced attention mechanisms (Wang and Jiang, 2016; Wang et al., 2017) might help hierarchical matching.”
Lightweight and Efficient Neural Natural Language Processing with Quaternion Networks (2019, methodology): “We use Tweets (Lan et al., 2017) data set and the Quora paraphrase data set (Wang et al., 2017).”
Beyond Fully-Connected Layers with Quaternions: Parameterization of Hypercomplex Multiplications with 1/n Parameters (2021, methodology): “, 2017), (ii) QNLI (Quora) (Wang et al., 2017), (iii) SNLI (Bowman et al.”
Multi-task Learning for Universal Sentence Embeddings: A Thorough Evaluation using Transfer and Auxiliary Tasks (2018, methodology): “We split the Quora dataset as that in Wang et al. (2017).”
Word Embedding based Edit Distance (2018, methodology): “We use the split of training, development, and test sets from previous work (Wang et al., 2017).”
A Survey on Text Classification: From Traditional to Deep Learning (2020, neutral): “[182] propose a model for the NLI task named Bilateral Multi-Perspective Matching (BiMPM).”
Neural Natural Language Inference Models Enhanced with External Knowledge (2017, neutral): “(Parikh et al., 2016) 86.8 NTI (Yu and Munkhdalai, 2017b) 87.3 Re-read LSTM (Sha et al., 2016) 87.5 BiMPM (Wang et al., 2017) 87.5 DIIN (Gong et al., 2017) 88.0 BCN + CoVe (McCann et al., 2017) 88.1 CAFE (Tay et al., 2018) 88.5 ESIM (Chen et al., 2017a) 88.0 KIM (This paper) 88.6 Table 2: Accuracies”
A Deep Look into Neural Ranking Models for Information Retrieval (2019, background): “A variety of neural ranking models [68, 18, 69, 70, 25] have been tested on these”