Joined Type Length Encoding for Nested Named Entity Recognition

In this article, we propose a new encoding scheme for named entity recognition (NER) called Joined Type-Length encoding (JoinedTL). Unlike most existing named entity encoding schemes, which focus on flat entities, JoinedTL can label nested named entities in a single sequence. JoinedTL uses a packed encoding to represent both type and span of a named entity, which not only results in less tagged tokens compared to existing encoding schemes, but also enables it to support nested NER. We evaluate the effectiveness of JoinedTL for nested NER on three nested NER datasets: GENIA in English, GermEval in German, and PerNest, our newly created nested NER dataset in Persian. We apply CharLSTM+WordLSTM+CRF, a three-layer sequence tagging model on three datasets encoded using JoinedTL and two existing nested NE encoding schemes, i.e., JoinedBIO and JoinedBILOU. Our experiment results show that CharLSTM+WordLSTM+CRF trained with JoinedTL encoded datasets can achieve competitive F1 scores as the ones trained with datasets encoded by two other encodings, but with 27%–48% less tagged tokens. To leverage the power of three different encodings, i.e., JoinedTL, JoinedBIO, and JoinedBILOU, we propose an encoding-based ensemble method for nested NER. Evaluation results show that the ensemble method achieves higher F1 scores on all datasets than the three models each trained using one of the three encodings. By using nested NE encodings including JoinedTL with CharLSTM+WordLSTM+CRF, we establish new state-of-the-art performance with an F1 score of 83.7 on PerNest, 74.9 on GENIA, and 70.5 on GermEval, surpassing two recent neural models specially designed for nested NER.

A Self-Supervised Representation Learning of Sentence Structure for Authorship Attribution

The syntactic structure of sentences in a document substantially informs about its authorial writing style. Sentence representation learning has been widely explored in recent years and it has been shown that it improves the generalization of different downstream tasks across many domains. Even though utilizing probing methods in several studies suggests that these learned contextual representations implicitly encode some amount of syntax, explicit syntactic information further improves the performance of deep neural models in the domain of authorship attribution. These observations have motivated us to investigate the explicit representation learning of syntactic structure of sentences. In this article, we propose a self-supervised framework for learning structural representations of sentences. The self-supervised network contains two components; a lexical sub-network and a syntactic sub-network which take the sequence of words and their corresponding structural labels as the input, respectively. Due to the n -to-1 mapping of words to their structural labels, each word will be embedded into a vector representation which mainly carries structural information. We evaluate the learned structural representations of sentences using different probing tasks, and subsequently utilize them in the authorship attribution task. Our experimental results indicate that the structural embeddings significantly improve the classification tasks when concatenated with the existing pre-trained word embeddings.

Time-Domain Joint Training Strategies of Speech Enhancement and Intent Classification Neural Models

Robustness against background noise and reverberation is essential for many real-world speech-based applications. One way to achieve this robustness is to employ a speech enhancement front-end that, independently of the back-end, removes the environmental perturbations from the target speech signal. However, although the enhancement front-end typically increases the speech quality from an intelligibility perspective, it tends to introduce distortions which deteriorate the performance of subsequent processing modules. In this paper, we investigate strategies for jointly training neural models for both speech enhancement and the back-end, which optimize a combined loss function. In this way, the enhancement front-end is guided by the back-end to provide more effective enhancement. Differently from typical state-of-the-art approaches employing on spectral features or neural embeddings, we operate in the time domain, processing raw waveforms in both components. As application scenario we consider intent classification in noisy environments. In particular, the front-end speech enhancement module is based on Wave-U-Net while the intent classifier is implemented as a temporal convolutional network. Exhaustive experiments are reported on versions of the Fluent Speech Commands corpus contaminated with noises from the Microsoft Scalable Noisy Speech Dataset, shedding light and providing insight about the most promising training approaches.

Biomaterial-based strategies for in vitro neural models

This review provides a comprehensive compendium of commonly used biomaterials as well as the different fabrication techniques employed for the design of 3D neural tissue models.

Morphological Analysis of Japanese Hiragana Sentences using the BI-LSTM CRF Model

This study proposes a method to develop neural models of the morphological analyzer for Japanese Hiragana sentences using the Bi-LSTM CRF model. Morphological analysis is a technique that divides text data into words and assigns information such as parts of speech. This technique plays an essential role in downstream applications in Japanese natural language processing systems because the Japanese language does not have word delimiters between words. Hiragana is a type of Japanese phonogramic characters, which is used for texts for children or people who cannot read Chinese characters. Morphological analysis of Hiragana sentences is more difficult than that of ordinary Japanese sentences because there is less information for dividing. For morphological analysis of Hiragana sentences, we demonstrated the effectiveness of fine-tuning using a model based on ordinary Japanese text and examined the influence of training data on texts of various genres.

Deep Learning for Text Style Transfer: A Survey

Text style transfer is an important task in natural language generation, which aims to control certain attributes in the generated text, such as politeness, emotion, humor, and many others. It has a long history in the field of natural language processing, and recently has re-gained significant attention thanks to the promising performance brought by deep neural models. In this paper, we present a systematic survey of the research on neural text style transfer, spanning over 100 representative articles since the first neural text style transfer work in 2017. We discuss the task formulation, existing datasets and subtasks, evaluation, as well as the rich methodologies in the presence of parallel and non-parallel data. We also provide discussions on a variety of important topics regarding the future development of this task.

A Probabilistic Re-Intepretation of Confidence Scores in Multi-Exit Models

In this paper, we propose a new approach to train a deep neural network with multiple intermediate auxiliary classifiers, branching from it. These ‘multi-exits’ models can be used to reduce the inference time by performing early exit on the intermediate branches, if the confidence of the prediction is higher than a threshold. They rely on the assumption that not all the samples require the same amount of processing to yield a good prediction. In this paper, we propose a way to train jointly all the branches of a multi-exit model without hyper-parameters, by weighting the predictions from each branch with a trained confidence score. Each confidence score is an approximation of the real one produced by the branch, and it is calculated and regularized while training the rest of the model. We evaluate our proposal on a set of image classification benchmarks, using different neural models and early-exit stopping criteria.