GPT-3 has attracted lots of attention due to its superior performance across a wide range of NLP tasks, especially with its in-context learning abilities. Despite its success, we found that the empirical results of GPT-3 depend heavily on the choice of in-context examples. In this work, we investigate whether there are more effective strategies for judiciously selecting in-context examples (relative to random sampling) that better leverage GPT-3’s in-context learning capabilities. Inspired by the recent success of leveraging a retrieval module to augment neural networks, we propose to retrieve examples that are semantically-similar to a test query sample to formulate its corresponding prompt. Intuitively, the examples selected with such a strategy may serve as more informative inputs to unleash GPT-3’s power of text generation. We evaluate the proposed approach on several natural language understanding and generation benchmarks, where the retrieval-based prompt selection approach consistently outperforms the random selection baseline. Moreover, it is observed that the sentence encoders fine-tuned on task-related datasets yield even more helpful retrieval results. Notably, significant gains are observed on tasks such as table-to-text generation (44.3% on the ToTTo dataset) and open-domain question answering (45.5% on the NQ dataset).

Open world classification is a task in natural language processing with key practical relevance and impact. Since the open or unknown category data only manifests in the inference phase, finding a model with a suitable decision boundary accommodating for the identification of known classes and discrimination of the open category is challenging. The performance of existing models is limited by the lack of effective open category data during the training stage or the lack of a good mechanism to learn appropriate decision boundaries. We propose an approach based on Adaptive Negative Samples (ANS) designed to generate effective synthetic open category samples in the training stage and without requiring any prior knowledge or external datasets. Empirically, we find a significant advantage in using auxiliary one-versus-rest binary classifiers, which effectively utilize the generated negative samples and avoid the complex threshold-seeking stage in previous works. Extensive experiments on three benchmark datasets show that ANS achieves significant improvements over state-of-the-art methods.

Natural language often exhibits inherent hierarchical structure ingrained with complex syntax and semantics. However, most state-of-the-art deep generative models learn embeddings only in Euclidean vector space, without accounting for this structural property of language. In this paper, we investigate text generation in a hyperbolic latent space to learn continuous hierarchical representations. An Adversarial Poincare Variational Autoencoder (APo-VAE) is presented, where both the prior and variational posterior of latent variables are defined over a Poincare ball via wrapped normal distributions. By adopting the primal-dual formulation of Kullback-Leibler divergence, an adversarial learning procedure is introduced to empower robust model training. Extensive experiments in language modeling, unaligned style transfer, and dialog-response generation demonstrate the effectiveness of the proposed APo-VAE model over VAEs in Euclidean latent space, thanks to its superb capabilities in capturing latent language hierarchies in hyperbolic space.

In many natural language processing applications, identifying predictive text can be as important as the predictions themselves. When predicting medical diagnoses, for example, identifying predictive content in clinical notes not only enhances interpretability, but also allows unknown, descriptive (i.e., text-based) risk factors to be identified. We here formalize this problem as predictive extraction and address it using a simple mechanism based on linear attention. Our method preserves differentiability, allowing scalable inference via stochastic gradient descent. Further, the model decomposes predictions into a sum of contributions of distinct text spans. Importantly, we require only document labels, not ground-truth spans. Results show that our model identifies semantically-cohesive spans and assigns them scores that agree with human ratings, while preserving classification performance.

It has been shown that training multi-task models with auxiliary tasks can improve the target task quality through cross-task transfer. However, the importance of each auxiliary task to the primary task is likely not known a priori. While the importance weights of auxiliary tasks can be manually tuned, it becomes practically infeasible with the number of tasks scaling up. To address this, we propose a search method that automatically assigns importance weights. We formulate it as a reinforcement learning problem and learn a task sampling schedule based on the evaluation accuracy of the multi-task model. Our empirical evaluation on XNLI and GLUE shows that our method outperforms uniform sampling and the corresponding single-task baseline.

In sequence-to-sequence models, classical optimal transport (OT) can be applied to semantically match generated sentences with target sentences. However, in non-parallel settings, target sentences are usually unavailable. To tackle this issue without losing the benefits of classical OT, we present a semantic matching scheme based on the Optimal Partial Transport (OPT). Specifically, our approach partially matches semantically meaningful words between source and partial target sequences. To overcome the difficulty of detecting active regions in OPT (corresponding to the words needed to be matched), we further exploit prior knowledge to perform partial matching. Extensive experiments are conducted to evaluate the proposed approach, showing consistent improvements over sequence-to-sequence tasks.

Pretrained Language Models (PLMs) have improved the performance of natural language understanding in recent years. Such models are pretrained on large corpora, which encode the general prior knowledge of natural languages but are agnostic to information characteristic of downstream tasks. This often results in overfitting when fine-tuned with low resource datasets where task-specific information is limited. In this paper, we integrate label information as a task-specific prior into the self-attention component of pretrained BERT models. Experiments on several benchmarks and real-word datasets suggest that the proposed approach can largely improve the performance of pretrained models when fine-tuning with small datasets.

Auto-regressive text generation models usually focus on local fluency, and may cause inconsistent semantic meaning in long text generation. Further, automatically generating words with similar semantics is challenging, and hand-crafted linguistic rules are difficult to apply. We consider a text planning scheme and present a model-based imitation-learning approach to alleviate the aforementioned issues. Specifically, we propose a novel guider network to focus on the generative process over a longer horizon, which can assist next-word prediction and provide intermediate rewards for generator optimization. Extensive experiments demonstrate that the proposed method leads to improved performance.

Learning disentangled representations of natural language is essential for many NLP tasks, e.g., conditional text generation, style transfer, personalized dialogue systems, etc. Similar problems have been studied extensively for other forms of data, such as images and videos. However, the discrete nature of natural language makes the disentangling of textual representations more challenging (e.g., the manipulation over the data space cannot be easily achieved). Inspired by information theory, we propose a novel method that effectively manifests disentangled representations of text, without any supervision on semantics. A new mutual information upper bound is derived and leveraged to measure dependence between style and content. By minimizing this upper bound, the proposed method induces style and content embeddings into two independent low-dimensional spaces. Experiments on both conditional text generation and text-style transfer demonstrate the high quality of our disentangled representation in terms of content and style preservation.

Legislator preferences are typically represented as measures of general ideology estimated from roll call votes on legislation, potentially masking important nuances in legislators’ political attitudes. In this paper we introduce a method of measuring more specific legislator attitudes using an alternative expression of preferences: tweeting. Specifically, we present an embedding-based model for predicting the frequency and sentiment of legislator tweets. To illustrate our method, we model legislators’ attitudes towards President Donald Trump as vector embeddings that interact with embeddings for Trump himself constructed using a neural network from the text of his daily tweets. We demonstrate the predictive performance of our model on tweets authored by members of the U.S. House and Senate related to the president from November 2016 to February 2018. We further assess the quality of our learned representations for legislators by comparing to traditional measures of legislator preferences.

Word embedding models are typically able to capture the semantics of words via the distributional hypothesis, but fail to capture the numerical properties of numbers that appear in the text. This leads to problems with numerical reasoning involving tasks such as question answering. We propose a new methodology to assign and learn embeddings for numbers. Our approach creates Deterministic, Independent-of-Corpus Embeddings (the model is referred to as DICE) for numbers, such that their cosine similarity reflects the actual distance on the number line. DICE outperforms a wide range of pre-trained word embedding models across multiple examples of two tasks: (i) evaluating the ability to capture numeration and magnitude; and (ii) to perform list maximum, decoding, and addition. We further explore the utility of these embeddings in downstream tasks, by initializing numbers with our approach for the task of magnitude prediction. We also introduce a regularization approach to learn model-based embeddings of numbers in a contextual setting.

Neural language models are often trained with maximum likelihood estimation (MLE), where the next word is generated conditioned on the ground-truth word tokens. During testing, however, the model is instead conditioned on previously generated tokens, resulting in what is termed exposure bias. To reduce this gap between training and testing, we propose using optimal transport (OT) to match the sequences generated in these two modes. We examine the necessity of adding Student-Forcing scheme during training with an imitation learning interpretation. An extension is further proposed to improve the OT learning for long sequences, based on the structural and contextual information of the text sequences. The effectiveness of the proposed method is validated on machine translation, text summarization, and text generation tasks.

We propose a topic-guided variational auto-encoder (TGVAE) model for text generation. Distinct from existing variational auto-encoder (VAE) based approaches, which assume a simple Gaussian prior for latent code, our model specifies the prior as a Gaussian mixture model (GMM) parametrized by a neural topic module. Each mixture component corresponds to a latent topic, which provides a guidance to generate sentences under the topic. The neural topic module and the VAE-based neural sequence module in our model are learned jointly. In particular, a sequence of invertible Householder transformations is applied to endow the approximate posterior of the latent code with high flexibility during the model inference. Experimental results show that our TGVAE outperforms its competitors on both unconditional and conditional text generation, which can also generate semantically-meaningful sentences with various topics.

Variational autoencoders (VAE) with an auto-regressive decoder have been applied for many natural language processing (NLP) tasks. VAE objective consists of two terms, the KL regularization term and the reconstruction term, balanced by a weighting hyper-parameter 𝛽. One notorious training difficulty is that the KL term tends to vanish. In this paper we study different scheduling schemes for 𝛽, and show that KL vanishing is caused by the lack of good latent codes in training decoder at the beginning of optimization. To remedy the issue, we propose a cyclical annealing schedule, which simply repeats the process of increasing 𝛽 multiple times. This new procedure allows us to learn more meaningful latent codes progressively by leveraging the results of previous learning cycles as warm re-restart. The effectiveness of cyclical annealing schedule is validated on a broad range of NLP tasks, including language modeling, dialog response generation and semi-supervised text classification.

Generating high-quality paraphrases is a fundamental yet challenging natural language processing task. Despite the effectiveness of previous work based on generative models, there remain problems with exposure bias in recurrent neural networks, and often a failure to generate realistic sentences. To overcome these challenges, we propose the first end-to-end conditional generative architecture for generating paraphrases via adversarial training, which does not depend on extra linguistic information. Extensive experiments on four public datasets demonstrate the proposed method achieves state-of-the-art results, outperforming previous generative architectures on both automatic metrics (BLEU, METEOR, and TER) and human evaluations.

Vector representations of sentences, trained on massive text corpora, are widely used as generic sentence embeddings across a variety of NLP problems. The learned representations are generally assumed to be continuous and real-valued, giving rise to a large memory footprint and slow retrieval speed, which hinders their applicability to low-resource (memory and computation) platforms, such as mobile devices. In this paper, we propose four different strategies to transform continuous and generic sentence embeddings into a binarized form, while preserving their rich semantic information. The introduced methods are evaluated across a wide range of downstream tasks, where the binarized sentence embeddings are demonstrated to degrade performance by only about 2% relative to their continuous counterparts, while reducing the storage requirement by over 98%. Moreover, with the learned binary representations, the semantic relatedness of two sentences can be evaluated by simply calculating their Hamming distance, which is more computational efficient compared with the inner product operation between continuous embeddings. Detailed analysis and case study further validate the effectiveness of proposed methods.

We present a syntax-infused variational autoencoder (SIVAE), that integrates sentences with their syntactic trees to improve the grammar of generated sentences. Distinct from existing VAE-based text generative models, SIVAE contains two separate latent spaces, for sentences and syntactic trees. The evidence lower bound objective is redesigned correspondingly, by optimizing a joint distribution that accommodates two encoders and two decoders. SIVAE works with long short-term memory architectures to simultaneously generate sentences and syntactic trees. Two versions of SIVAE are proposed: one captures the dependencies between the latent variables through a conditional prior network, and the other treats the latent variables independently such that syntactically-controlled sentence generation can be performed. Experimental results demonstrate the generative superiority of SIVAE on both reconstruction and targeted syntactic evaluations. Finally, we show that the proposed models can be used for unsupervised paraphrasing given different syntactic tree templates.

Variational autoencoders (VAEs) have received much attention recently as an end-to-end architecture for text generation with latent variables. However, previous works typically focus on synthesizing relatively short sentences (up to 20 words), and the posterior collapse issue has been widely identified in text-VAEs. In this paper, we propose to leverage several multi-level structures to learn a VAE model for generating long, and coherent text. In particular, a hierarchy of stochastic layers between the encoder and decoder networks is employed to abstract more informative and semantic-rich latent codes. Besides, we utilize a multi-level decoder structure to capture the coherent long-term structure inherent in long-form texts, by generating intermediate sentence representations as high-level plan vectors. Extensive experimental results demonstrate that the proposed multi-level VAE model produces more coherent and less repetitive long text compared to baselines as well as can mitigate the posterior-collapse issue.

Constituting highly informative network embeddings is an essential tool for network analysis. It encodes network topology, along with other useful side information, into low dimensional node-based feature representations that can be exploited by statistical modeling. This work focuses on learning context-aware network embeddings augmented with text data. We reformulate the network embedding problem, and present two novel strategies to improve over traditional attention mechanisms: (i) a content-aware sparse attention module based on optimal transport; and (ii) a high-level attention parsing module. Our approach yields naturally sparse and self-normalized relational inference. It can capture long-term interactions between sequences, thus addressing the challenges faced by existing textual network embedding schemes. Extensive experiments are conducted to demonstrate our model can consistently outperform alternative state-of-the-art methods.

Network embeddings, which learns low-dimensional representations for each vertex in a large-scale network, have received considerable attention in recent years. For a wide range of applications, vertices in a network are typically accompanied by rich textual information such as user profiles, paper abstracts, etc. In this paper, we propose to incorporate semantic features into network embeddings by matching important words between text sequences for all pairs of vertices. We introduce an word-by-word alignment framework that measures the compatibility of embeddings between word pairs, and then adaptively accumulates these alignment features with a simple yet effective aggregation function. In experiments, we evaluate the proposed framework on three real-world benchmarks for downstream tasks, including link prediction and multi-label vertex classification. The experimental results demonstrate that our model outperforms state-of-the-art network embedding methods by a large margin.

Convolutional neural networks (CNNs) have recently emerged as a popular building block for natural language processing (NLP). Despite their success, most existing CNN models employed in NLP share the same learned (and static) set of filters for all input sentences. In this paper, we consider an approach of using a small meta network to learn context-sensitive convolutional filters for text processing. The role of meta network is to abstract the contextual information of a sentence or document into a set of input-sensitive filters. We further generalize this framework to model sentence pairs, where a bidirectional filter generation mechanism is introduced to encapsulate co-dependent sentence representations. In our benchmarks on four different tasks, including ontology classification, sentiment analysis, answer sentence selection, and paraphrase identification, our proposed model, a modified CNN with context-sensitive filters, consistently outperforms the standard CNN and attention-based CNN baselines. By visualizing the learned context-sensitive filters, we further validate and rationalize the effectiveness of proposed framework.

Many deep learning architectures have been proposed to model the compositionality in text sequences, requiring substantial number of parameters and expensive computations. However, there has not been a rigorous evaluation regarding the added value of sophisticated compositional functions. In this paper, we conduct a point-by-point comparative study between Simple Word-Embedding-based Models (SWEMs), consisting of parameter-free pooling operations, relative to word-embedding-based RNN/CNN models. Surprisingly, SWEMs exhibit comparable or even superior performance in the majority of cases considered. Based upon this understanding, we propose two additional pooling strategies over learned word embeddings: (i) a max-pooling operation for improved interpretability; and (ii) a hierarchical pooling operation, which preserves spatial (n-gram) information within text sequences. We present experiments on 17 datasets encompassing three tasks: (i) (long) document classification; (ii) text sequence matching; and (iii) short text tasks, including classification and tagging.

Semantic hashing has become a powerful paradigm for fast similarity search in many information retrieval systems. While fairly successful, previous techniques generally require two-stage training, and the binary constraints are handled ad-hoc. In this paper, we present an end-to-end Neural Architecture for Semantic Hashing (NASH), where the binary hashing codes are treated as Bernoulli latent variables. A neural variational inference framework is proposed for training, where gradients are directly backpropagated through the discrete latent variable to optimize the hash function. We also draw the connections between proposed method and rate-distortion theory, which provides a theoretical foundation for the effectiveness of our framework. Experimental results on three public datasets demonstrate that our method significantly outperforms several state-of-the-art models on both unsupervised and supervised scenarios.

Word embeddings are effective intermediate representations for capturing semantic regularities between words, when learning the representations of text sequences. We propose to view text classification as a label-word joint embedding problem: each label is embedded in the same space with the word vectors. We introduce an attention framework that measures the compatibility of embeddings between text sequences and labels. The attention is learned on a training set of labeled samples to ensure that, given a text sequence, the relevant words are weighted higher than the irrelevant ones. Our method maintains the interpretability of word embeddings, and enjoys a built-in ability to leverage alternative sources of information, in addition to input text sequences. Extensive results on the several large text datasets show that the proposed framework outperforms the state-of-the-art methods by a large margin, in terms of both accuracy and speed.

We propose a new encoder-decoder approach to learn distributed sentence representations that are applicable to multiple purposes. The model is learned by using a convolutional neural network as an encoder to map an input sentence into a continuous vector, and using a long short-term memory recurrent neural network as a decoder. Several tasks are considered, including sentence reconstruction and future sentence prediction. Further, a hierarchical encoder-decoder model is proposed to encode a sentence to predict multiple future sentences. By training our models on a large collection of novels, we obtain a highly generic convolutional sentence encoder that performs well in practice. Experimental results on several benchmark datasets, and across a broad range of applications, demonstrate the superiority of the proposed model over competing methods.

Recurrent neural networks (RNNs) have shown promising performance for language modeling. However, traditional training of RNNs using back-propagation through time often suffers from overfitting. One reason for this is that stochastic optimization (used for large training sets) does not provide good estimates of model uncertainty. This paper leverages recent advances in stochastic gradient Markov Chain Monte Carlo (also appropriate for large training sets) to learn weight uncertainty in RNNs. It yields a principled Bayesian learning algorithm, adding gradient noise during training (enhancing exploration of the model-parameter space) and model averaging when testing. Extensive experiments on various RNN models and across a broad range of applications demonstrate the superiority of the proposed approach relative to stochastic optimization.