Analyzing Linguistic Knowledge in Sequential Model of Sentence

By Peng Qian, Xipeng Qiu, Xuanjing Huang. 2016

36 citations recorded by the source corpus.

Sentence modelling is a fundamental topic in computational linguistics. Recently, deep learning-based sequential models of sentence, such as recurrent neural network, have proved to be effective in dealing with the non-sequential properties of human language. However, little is known about how a recurrent neural network captures linguistic knowledge. Here we propose to correlate the neuron activation pattern of a LSTM language model with rich language features at sequential, lexical and compositional level. Qualitative visualization as well as quantitative analysis under multilingual perspective reveals the effectiveness of gate neurons and indicates that LSTM learns to allow different neurons selectively respond to linguistic knowledge at different levels. Cross-language evidence shows that the model captures different aspects of linguistic properties for different languages due to the variance of syntactic complexity. Additionally, we analyze the inﬂuence of modelling strategy on linguistic knowledge encoded implicitly in different sequential models.

Subfields: Artificial Intelligence, Natural Language Processing, Computational Linguistics & Semantics

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 Computational Linguistics & Semantics

Citations in the corpus

Analysis Methods in Neural Language Processing: A Survey (2018, background): “2015; Kádár et al., 2017; Qian et al., 2016a; Liu et al., 2018) and speech (Wu and King, 2016; Nagamine et al., 2015; Wang et al., 2017b).”
What do Neural Machine Translation Models Learn about Morphology? (2017, methodology): “Different units have been used, from word embeddings (K¨ohn, 2015; Qian et al., 2016b), through LSTM gates or states (Qian et al., 2016a), to sentence embeddings (Adi et al., 2016).”
Evaluating Layers of Representation in Neural Machine Translation on Part-of-Speech and Semantic Tagging Tasks (2017, methodology): “Techniques for analyzing neural network models include visualization of hidden units (Elman, 1991; Karpathy et al., 2015; Kádár et al., 2016; Qian et al., 2016a), which provide illuminating, but often anecdotal information on how the network works.”
An Analysis of Encoder Representations in Transformer-Based Machine Translation (2018, background): “Other recent studies look for hidden linguistic units that provide information on how the network works (Karpathy et al., 2015; Qian et al., 2016; Kádár et al., 2017), while another line of analysis probes the representation learned by a neural network as input to a classifier of another task…”
Identifying and Controlling Important Neurons in Neural Machine Translation (2018, background): “In contrast, previous work found meaningful individual neurons in computer vision (Zeiler & Fergus, 2014; Zhou et al., 2016; Bau et al., 2017, among others) and in a few NLP tasks (Karpathy et al., 2015; Radford et al., 2017; Qian et al., 2016a).”
One Size Does Not Fit All: Comparing NMT Representations of Different Granularities (2019, methodology): “…analyzing word and sentence embeddings (Adi et al., 2017; Qian et al., 2016b; Ganesh et al., 2017; Conneau et al., 2018, among others), RNN states (Qian et al., 2016a; Shi et al., 2016; Wu and King, 2016; Wang et al., 2017), and NMT representations (Shi et al., 2016; Belinkov et al., 2017a), as…”
Analyzing Individual Neurons in Pre-trained Language Models (2020, unknown): “Linear classiﬁers are a popular choice in analyzing deep NLP models due to their better interpretability (Qian et al., 2016; Belinkov et al., 2020).”
What Is One Grain of Sand in the Desert? Analyzing Individual Neurons in Deep NLP Models (2018, background): “Some recent work on neural language models and machine translation analyzes speciﬁc neurons of length (Qian, Qiu, and Huang 2016a; Shi, Knight, and Yuret 2016) and sentiment (Radford, Jozefowicz, and Sutskever 2017).”
Analyzing Redundancy in Pretrained Transformer Models (2020, background): “Linear classifiers are a popular choice in analyzing deep NLP models due to their better interpretability (Qian et al., 2016; Belinkov et al., 2020).”
On internal language representations in deep learning: an analysis of machine translation and speech recognition (2018, background): “Much recent work has focused on visualizing activations on specific examples in modern neural networks for language [103, 168, 173, 278] and speech [251, 345].”
Similarity Analysis of Contextual Word Representation Models (2020, background): “While neural language representations are thought to be densely distributed, several recent studies have pointed out the importance of individual neurons (Qian et al., 2016; Shi et al., 2016; Radford et al., 2017; Lakretz et al., 2019; Bau et al., 2019; Dalvi et al., 2019; Baan et al., 2019).”