相关论文: Inducing Syntactic Trees from BERT Representations

We present iBERT (interpretable-BERT), an encoder to produce inherently interpretable and controllable embeddings - designed to modularize and expose the discriminative cues present in language, such as semantic or stylistic structure. Each…

How does word frequency in pre-training data affect the behavior of similarity metrics in contextualized BERT embeddings? Are there systematic ways in which some word relationships are exaggerated or understated? In this work, we explore…

Pretrained language models have achieved a new state of the art on many NLP tasks, but there are still many open questions about how and why they work so well. We investigate the contextualization of words in BERT. We quantify the amount of…

Pretraining deep language models has led to large performance gains in NLP. Despite this success, Schick and Sch\"utze (2020) recently showed that these models struggle to understand rare words. For static word embeddings, this problem has…

In this paper we introduce Latent Tree Language Model (LTLM), a novel approach to language modeling that encodes syntax and semantics of a given sentence as a tree of word roles. The learning phase iteratively updates the trees by moving…

The capabilities and limitations of BERT and similar models are still unclear when it comes to learning syntactic abstractions, in particular across languages. In this paper, we use the task of subordinate-clause detection within and across…

We evaluate whether BERT, a widely used neural network for sentence processing, acquires an inductive bias towards forming structural generalizations through pretraining on raw data. We conduct four experiments testing its preference for…

Tremendous amounts of multimedia associated with speech information are driving an urgent need to develop efficient and effective automatic summarization methods. To this end, we have seen rapid progress in applying supervised deep neural…

Active learning has been shown to be an effective way to alleviate some of the effort required in utilising large collections of unlabelled data for machine learning tasks without needing to fully label them. The representation mechanism…

Domain adaptation or transfer learning using pre-trained language models such as BERT has proven to be an effective approach for many natural language processing tasks. In this work, we propose to formulate word sense disambiguation as a…

Contextual word embeddings obtained from pre-trained language model (PLM) have proven effective for various natural language processing tasks at the word level. However, interpreting the hidden aspects within embeddings, such as syntax and…

We apply decision tree induction to the problem of discourse clue word sense disambiguation with a genetic algorithm. The automatic partitioning of the training set which is intrinsic to decision tree induction gives rise to linguistically…

We use reinforcement learning to learn tree-structured neural networks for computing representations of natural language sentences. In contrast with prior work on tree-structured models in which the trees are either provided as input or…

Large scale language models encode rich commonsense knowledge acquired through exposure to massive data during pre-training, but their understanding of entities and their semantic properties is unclear. We probe BERT (Devlin et al., 2019)…

Measuring the quality of a generated sequence against a set of references is a central problem in many learning frameworks, be it to compute a score, to assign a reward, or to perform discrimination. Despite great advances in model…

Lexical simplification (LS) aims to replace complex words in a given sentence with their simpler alternatives of equivalent meaning. Recently unsupervised lexical simplification approaches only rely on the complex word itself regardless of…

We propose a generative model for a sentence that uses two latent variables, with one intended to represent the syntax of the sentence and the other to represent its semantics. We show we can achieve better disentanglement between semantic…

Syntax is a latent hierarchical structure which underpins the robust and compositional nature of human language. In this work, we explore the hypothesis that syntactic dependencies can be represented in language model attention…

Do state-of-the-art natural language understanding models care about word order - one of the most important characteristics of a sequence? Not always! We found 75% to 90% of the correct predictions of BERT-based classifiers, trained on many…

We introduce a neural network that represents sentences by composing their words according to induced binary parse trees. We use Tree-LSTM as our composition function, applied along a tree structure found by a fully differentiable natural…