Large pretrained language models (LMs) like BERT have improved performance in many disparate natural language processing (NLP) tasks. However, fine tuning such models requires a large number of training examples for each target task. Simultaneously, many realistic NLP problems are "few shot", without a sufficiently large training set. In this work, we propose a novel conditional neural process-based approach for few-shot text classification that learns to transfer from other diverse tasks with rich annotation. Our key idea is to represent each task using gradient information from a base model and to train an adaptation network that modulates a text classifier conditioned on the task representation. While previous task-aware few-shot learners represent tasks by input encoding, our novel task representation is more powerful, as the gradient captures input-output relationships of a task. Experimental results show that our approach outperforms traditional fine-tuning, sequential transfer learning, and state-of-the-art meta learning approaches on a collection of diverse few-shot tasks. We further conducted analysis and ablations to justify our design choices.
An overwhelmingly large amount of knowledge in the materials domain is generated and stored as text published in peer-reviewed scientific literature. Recent developments in natural language processing, such as bidirectional encoder representations from transformers (BERT) models, provide promising tools to extract information from these texts. However, direct application of these models in the materials domain may yield suboptimal results as the models themselves may not be trained on notations and jargon that are specific to the domain. Here, we present a materials-aware language model, namely, MatSciBERT, which is trained on a large corpus of scientific literature published in the materials domain. We further evaluate the performance of MatSciBERT on three downstream tasks, namely, abstract classification, named entity recognition, and relation extraction, on different materials datasets. We show that MatSciBERT outperforms SciBERT, a language model trained on science corpus, on all the tasks. Further, we discuss some of the applications of MatSciBERT in the materials domain for extracting information, which can, in turn, contribute to materials discovery or optimization. Finally, to make the work accessible to the larger materials community, we make the pretrained and finetuned weights and the models of MatSciBERT freely accessible.
Artificial Neural Network (ANN) s has widely been used for recognition of optically scanned character, which partially emulates human thinking in the domain of the Artificial Intelligence. But prior to recognition, it is necessary to segment the character from the text to sentences, words etc. Segmentation of words into individual letters has been one of the major problems in handwriting recognition. Despite several successful works all over the work, development of such tools in specific languages is still an ongoing process especially in the Indian context. This work explores the application of ANN as an aid to segmentation of handwritten characters in Assamese- an important language in the North Eastern part of India. The work explores the performance difference obtained in applying an ANN-based dynamic segmentation algorithm compared to projection- based static segmentation. The algorithm involves, first training of an ANN with individual handwritten characters recorded from different individuals. Handwritten sentences are separated out from text using a static segmentation method. From the segmented line, individual characters are separated out by first over segmenting the entire line. Each of the segments thus obtained, next, is fed to the trained ANN. The point of segmentation at which the ANN recognizes a segment or a combination of several segments to be similar to a handwritten character, a segmentation boundary for the character is assumed to exist and segmentation performed. The segmented character is next compared to the best available match and the segmentation boundary confirmed.
Previous works on expressive text-to-speech (TTS) have a limitation on robustness and speed when training and inferring. Such drawbacks mostly come from autoregressive decoding, which makes the succeeding step vulnerable to preceding error. To overcome this weakness, we propose STYLER, a novel expressive text-to-speech model with parallelized architecture. Expelling autoregressive decoding and introducing speech decomposition for encoding enables speech synthesis more robust even with high style transfer performance. Moreover, our novel noise modeling approach from audio using domain adversarial training and Residual Decoding enabled style transfer without transferring noise. Our experiments prove the naturalness and expressiveness of our model from comparison with other parallel TTS models. Together we investigate our model's robustness and speed by comparison with the expressive TTS model with autoregressive decoding.
End-to-end Speech-to-text Translation (E2E- ST), which directly translates source language speech to target language text, is widely useful in practice, but traditional cascaded approaches (ASR+MT) often suffer from error propagation in the pipeline. On the other hand, existing end-to-end solutions heavily depend on the source language transcriptions for pre-training or multi-task training with Automatic Speech Recognition (ASR). We instead propose a simple technique to learn a robust speech encoder in a self-supervised fashion only on the speech side, which can utilize speech data without transcription. This technique, termed Masked Acoustic Modeling (MAM), can also perform pre-training, for the first time, on any acoustic signals (including non-speech ones) without annotation. Compared with current state-of-the-art models on ST, our technique achieves +1.4 BLEU improvement without using transcriptions, and +1.2 BLEU using transcriptions. The pre-training of MAM with arbitrary acoustic signals also boosts the downstream speech-related tasks.
Most adversarial attack methods that are designed to deceive a text classifier change the text classifier's prediction by modifying a few words or characters. Few try to attack classifiers by rewriting a whole sentence, due to the difficulties inherent in sentence-level rephrasing as well as the problem of setting the criteria for legitimate rewriting. In this paper, we explore the problem of creating adversarial examples with sentence-level rewriting. We design a new sampling method, named ParaphraseSampler, to efficiently rewrite the original sentence in multiple ways. Then we propose a new criteria for modification, called a sentence-level threaten model. This criteria allows for both word- and sentence-level changes, and can be adjusted independently in two dimensions: semantic similarity and grammatical quality. Experimental results show that many of these rewritten sentences are misclassified by the classifier. On all 6 datasets, our ParaphraseSampler achieves a better attack success rate than our baseline.
This thesis presents a language-independent text classification model by introduced two new encoding methods "BUNOW" and "BUNOC" used for feeding the raw text data into a new CNN spatial architecture with vertical and horizontal convolutional process instead of commonly used methods like one hot vector or word representation (i.e. word2vec) with temporal CNN architecture. The proposed model can be classified as hybrid word-character model in its work methodology because it consumes less memory space by using a fewer neural network parameters as in character level representation, in addition to providing much faster computations with fewer network layers depth, as in word level representation. A promising result achieved compared to state of art models in two different morphological benchmarked dataset one for Arabic language and one for English language.
The knowledge extraction task is to extract triple relations (head entity-relation-tail entity) from unstructured text data. The existing knowledge extraction methods are divided into "pipeline" method and joint extraction method. The "pipeline" method is to separate named entity recognition and entity relationship extraction and use their own modules to extract them. Although this method has better flexibility, the training speed is slow. The learning model of joint extraction is an end-to-end model implemented by neural network to realize entity recognition and relationship extraction at the same time, which can well preserve the association between entities and relationships, and convert the joint extraction of entities and relationships into a sequence annotation problem. In this paper, we propose a knowledge extraction method for scientific and technological resources based on word mixture and GRU, combined with word mixture vector mapping method and self-attention mechanism, to effectively improve the effect of text relationship extraction for Chinese scientific and technological resources.
Fine-tuning pre-trained sentence embedding models like BERT has become the default transfer learning approach for several NLP tasks like text classification. We propose an alternative transfer learning approach called SimpleTran which is simple and effective for low resource text classification characterized by small sized datasets. We train a simple sentence embedding model on the target dataset, combine its output embedding with that of the pre-trained model via concatenation or dimension reduction, and finally train a classifier on the combined embedding either by fixing the embedding model weights or training the classifier and the embedding models end-to-end. Keeping embeddings fixed, SimpleTran significantly improves over fine-tuning on small datasets, with better computational efficiency. With end-to-end training, SimpleTran outperforms fine-tuning on small and medium sized datasets with negligible computational overhead. We provide theoretical analysis for our method, identifying conditions under which it has advantages.
Speech pre-training has primarily demonstrated efficacy on classification tasks, while its capability of generating novel speech, similar to how GPT-2 can generate coherent paragraphs, has barely been explored. Generative Spoken Language Modeling (GSLM) (Lakhotia et al., 2021) is the only prior work addressing the generative aspects of speech pre-training, which replaces text with discovered phone-like units for language modeling and shows the ability to generate meaningful novel sentences. Unfortunately, despite eliminating the need of text, the units used in GSLM discard most of the prosodic information. Hence, GSLM fails to leverage prosody for better comprehension, and does not generate expressive speech. In this work, we present a prosody-aware generative spoken language model (pGSLM). It is composed of a multi-stream transformer language model (MS-TLM) of speech, represented as discovered unit and prosodic feature streams, and an adapted HiFi-GAN model converting MS-TLM outputs to waveforms. We devise a series of metrics for prosody modeling and generation, and re-use metrics from GSLM for content modeling. Experimental results show that the pGSLM can utilize prosody to improve both prosody and content modeling, and also generate natural, meaningful, and coherent speech given a spoken prompt. Audio samples can be found at https://speechbot.github.io/pgslm.