Our entry into the HAHA 2019 Challenge placed $3^{rd}$ in the classification task and $2^{nd}$ in the regression task. We describe our system and innovations, as well as comparing our results to a Naive Bayes baseline. A large Twitter based corpus allowed us to train a language model from scratch focused on Spanish and transfer that knowledge to our competition model. To overcome the inherent errors in some labels we reduce our class confidence with label smoothing in the loss function. All the code for our project is included in a GitHub repository for easy reference and to enable replication by others.
We present a simple yet effective method for generating high quality classical Chinese poetry with Generative Pre-trained Language Model (GPT). The method adopts a simple GPT model, without using any human crafted rules or features, or designing any additional neural components. While the proposed model learns to generate various forms of classical Chinese poems, including Jueju, L\"{u}shi, various Cipai and Couples, the generated poems are of very high quality. We also propose and implement a method to fine-tune the model to generate acrostic poetry. To the best of our knowledge, this is the first to employ GPT in developing a poetry generation system. We have released an online mini demonstration program on Wechat to show the generation capability of the proposed method for classical Chinese poetry.
Pre-trained word embeddings are the primary method for transfer learning in several Natural Language Processing (NLP) tasks. Recent works have focused on using unsupervised techniques such as language modeling to obtain these embeddings. In contrast, this work focuses on extracting representations from multiple pre-trained supervised models, which enriches word embeddings with task and domain specific knowledge. Experiments performed in cross-task, cross-domain and cross-lingual settings indicate that such supervised embeddings are helpful, especially in the low-resource setting, but the extent of gains is dependent on the nature of the task and domain. We make our code publicly available.
We propose a contextual emotion classifier based on a transferable language model and dynamic max pooling, which predicts the emotion of each utterance in a dialogue. A representative emotion analysis task, EmotionX, requires to consider contextual information from colloquial dialogues and to deal with a class imbalance problem. To alleviate these problems, our model leverages the self-attention based transferable language model and the weighted cross entropy loss. Furthermore, we apply post-training and fine-tuning mechanisms to enhance the domain adaptability of our model and utilize several machine learning techniques to improve its performance. We conduct experiments on two emotion-labeled datasets named Friends and EmotionPush. As a result, our model outperforms the previous state-of-the-art model and also shows competitive performance in the EmotionX 2019 challenge. The code will be available in the Github page.
Latest development of neural models has connected the encoder and decoder through a self-attention mechanism. In particular, Transformer, which is solely based on self-attention, has led to breakthroughs in Natural Language Processing (NLP) tasks. However, the multi-head attention mechanism, as a key component of Transformer, limits the effective deployment of the model to a resource-limited setting. In this paper, based on the ideas of tensor decomposition and parameters sharing, we propose a novel self-attention model (namely Multi-linear attention) with Block-Term Tensor Decomposition (BTD). We test and verify the proposed attention method on three language modeling tasks (i.e., PTB, WikiText-103 and One-billion) and a neural machine translation task (i.e., WMT-2016 English-German). Multi-linear attention can not only largely compress the model parameters but also obtain performance improvements, compared with a number of language modeling approaches, such as Transformer, Transformer-XL, and Transformer with tensor train decomposition.
With the capability of modeling bidirectional contexts, denoising autoencoding based pretraining like BERT achieves better performance than pretraining approaches based on autoregressive language modeling. However, relying on corrupting the input with masks, BERT neglects dependency between the masked positions and suffers from a pretrain-finetune discrepancy. In light of these pros and cons, we propose XLNet, a generalized autoregressive pretraining method that (1) enables learning bidirectional contexts by maximizing the expected likelihood over all permutations of the factorization order and (2) overcomes the limitations of BERT thanks to its autoregressive formulation. Furthermore, XLNet integrates ideas from Transformer-XL, the state-of-the-art autoregressive model, into pretraining. Empirically, under comparable experiment settings, XLNet outperforms BERT on 20 tasks, often by a large margin, including question answering, natural language inference, sentiment analysis, and document ranking.
Recent advances in language modeling using deep neural networks have shown that these models learn representations, that vary with the network depth from morphology to semantic relationships like co-reference. We apply pre-trained language models to low-resource named entity recognition for Historic German. We show on a series of experiments that character-based pre-trained language models do not run into trouble when faced with low-resource datasets. Our pre-trained character-based language models improve upon classical CRF-based methods and previous work on Bi-LSTMs by boosting F1 score performance by up to 6%. Our pre-trained language and NER models are publicly available under https://github.com/stefan-it/historic-ner .
Modeling human language requires the ability to not only generate fluent text but also encode factual knowledge. However, traditional language models are only capable of remembering facts seen at training time, and often have difficulty recalling them. To address this, we introduce the knowledge graph language model (KGLM), a neural language model with mechanisms for selecting and copying facts from a knowledge graph that are relevant to the context. These mechanisms enable the model to render information it has never seen before, as well as generate out-of-vocabulary tokens. We also introduce the Linked WikiText-2 dataset, a corpus of annotated text aligned to the Wikidata knowledge graph whose contents (roughly) match the popular WikiText-2 benchmark. In experiments, we demonstrate that the KGLM achieves significantly better performance than a strong baseline language model. We additionally compare different language model's ability to complete sentences requiring factual knowledge, showing that the KGLM outperforms even very large language models in generating facts.
Architecture search is the process of automatically learning the neural model or cell structure that best suits the given task. Recently, this approach has shown promising performance improvements (on language modeling and image classification) with reasonable training speed, using a weight sharing strategy called Efficient Neural Architecture Search (ENAS). In our work, we first introduce a novel continual architecture search (CAS) approach, so as to continually evolve the model parameters during the sequential training of several tasks, without losing performance on previously learned tasks (via block-sparsity and orthogonality constraints), thus enabling life-long learning. Next, we explore a multi-task architecture search (MAS) approach over ENAS for finding a unified, single cell structure that performs well across multiple tasks (via joint controller rewards), and hence allows more generalizable transfer of the cell structure knowledge to an unseen new task. We empirically show the effectiveness of our sequential continual learning and parallel multi-task learning based architecture search approaches on diverse sentence-pair classification tasks (GLUE) and multimodal-generation based video captioning tasks. Further, we present several ablations and analyses on the learned cell structures.
In neural network models of language, words are commonly represented using context-invariant representations (word embeddings) which are then put in context in the hidden layers. Since words are often ambiguous, representing the contextually relevant information is not trivial. We investigate how an LSTM language model deals with lexical ambiguity in English, designing a method to probe its hidden representations for lexical and contextual information about words. We find that both types of information are represented to a large extent, but also that there is room for improvement for contextual information.
Large pre-trained neural networks such as BERT have had great recent success in NLP, motivating a growing body of research investigating what aspects of language they are able to learn from unlabeled data. Most recent analysis has focused on model outputs (e.g., language model surprisal) or internal vector representations (e.g., probing classifiers). Complementary to these works, we propose methods for analyzing the attention mechanisms of pre-trained models and apply them to BERT. BERT's attention heads exhibit patterns such as attending to delimiter tokens, specific positional offsets, or broadly attending over the whole sentence, with heads in the same layer often exhibiting similar behaviors. We further show that certain attention heads correspond well to linguistic notions of syntax and coreference. For example, we find heads that attend to the direct objects of verbs, determiners of nouns, objects of prepositions, and coreferent mentions with remarkably high accuracy. Lastly, we propose an attention-based probing classifier and use it to further demonstrate that substantial syntactic information is captured in BERT's attention.
Recently, substantial progress has been made in language modeling by using deep neural networks. However, in practice, large scale neural language models have been shown to be prone to overfitting. In this paper, we present a simple yet highly effective adversarial training mechanism for regularizing neural language models. The idea is to introduce adversarial noise to the output embedding layer while training the models. We show that the optimal adversarial noise yields a simple closed-form solution, thus allowing us to develop a simple and time efficient algorithm. Theoretically, we show that our adversarial mechanism effectively encourages the diversity of the embedding vectors, helping to increase the robustness of models. Empirically, we show that our method improves on the single model state-of-the-art results for language modeling on Penn Treebank (PTB) and Wikitext-2, achieving test perplexity scores of 46.01 and 38.07, respectively. When applied to machine translation, our method improves over various transformer-based translation baselines in BLEU scores on the WMT14 English-German and IWSLT14 German-English tasks.
Recently, there has been an increasing interest in unsupervised parsers that optimize semantically oriented objectives, typically using reinforcement learning. Unfortunately, the learned trees often do not match actual syntax trees well. Shen et al. (2018) propose a structured attention mechanism for language modeling (PRPN), which induces better syntactic structures but relies on ad hoc heuristics. Also, their model lacks interpretability as it is not grounded in parsing actions. In our work, we propose an imitation learning approach to unsupervised parsing, where we transfer the syntactic knowledge induced by the PRPN to a Tree-LSTM model with discrete parsing actions. Its policy is then refined by Gumbel-Softmax training towards a semantically oriented objective. We evaluate our approach on the All Natural Language Inference dataset and show that it achieves a new state of the art in terms of parsing $F$-score, outperforming our base models, including the PRPN.
Common language models typically predict the next word given the context. In this work, we propose a method that improves language modeling by learning to align the given context and the following phrase. The model does not require any linguistic annotation of phrase segmentation. Instead, we define syntactic heights and phrase segmentation rules, enabling the model to automatically induce phrases, recognize their task-specific heads, and generate phrase embeddings in an unsupervised learning manner. Our method can easily be applied to language models with different network architectures since an independent module is used for phrase induction and context-phrase alignment, and no change is required in the underlying language modeling network. Experiments have shown that our model outperformed several strong baseline models on different data sets. We achieved a new state-of-the-art performance of 17.4 perplexity on the Wikitext-103 dataset. Additionally, visualizing the outputs of the phrase induction module showed that our model is able to learn approximate phrase-level structural knowledge without any annotation.
Neural network architectures have been augmented with differentiable stacks in order to introduce a bias toward learning hierarchy-sensitive regularities. It has, however, proven difficult to assess the degree to which such a bias is effective, as the operation of the differentiable stack is not always interpretable. In this paper, we attempt to detect the presence of latent representations of hierarchical structure through an exploration of the unsupervised learning of constituency structure. Using a technique due to Shen et al. (2018a,b), we extract syntactic trees from the pushing behavior of stack RNNs trained on language modeling and classification objectives. We find that our models produce parses that reflect natural language syntactic constituencies, demonstrating that stack RNNs do indeed infer linguistically relevant hierarchical structure.
Introducing common sense to natural language understanding systems has received increasing research attention. It remains a fundamental question on how to evaluate whether a system has a sense making capability. Existing benchmarks measures commonsense knowledge indirectly and without explanation. In this paper, we release a benchmark to directly test whether a system can differentiate natural language statements that make sense from those that do not make sense. In addition, a system is asked to identify the most crucial reason why a statement does not make sense. We evaluate models trained over large-scale language modeling tasks as well as human performance, showing that there are different challenges for system sense making.
Medication recommendation is an important healthcare application. It is commonly formulated as a temporal prediction task. Hence, most existing works only utilize longitudinal electronic health records (EHRs) from a small number of patients with multiple visits ignoring a large number of patients with a single visit (selection bias). Moreover, important hierarchical knowledge such as diagnosis hierarchy is not leveraged in the representation learning process. To address these challenges, we propose G-BERT, a new model to combine the power of Graph Neural Networks (GNNs) and BERT (Bidirectional Encoder Representations from Transformers) for medical code representation and medication recommendation. We use GNNs to represent the internal hierarchical structures of medical codes. Then we integrate the GNN representation into a transformer-based visit encoder and pre-train it on EHR data from patients only with a single visit. The pre-trained visit encoder and representation are then fine-tuned for downstream predictive tasks on longitudinal EHRs from patients with multiple visits. G-BERT is the first to bring the language model pre-training schema into the healthcare domain and it achieved state-of-the-art performance on the medication recommendation task.
Tables contain valuable knowledge in a structured form. We employ neural language modeling approaches to embed tabular data into vector spaces. Specifically, we consider different table elements, such caption, column headings, and cells, for training word and entity embeddings. These embeddings are then utilized in three particular table-related tasks, row population, column population, and table retrieval, by incorporating them into existing retrieval models as additional semantic similarity signals. Evaluation results show that table embeddings can significantly improve upon the performance of state-of-the-art baselines.
Does adding a theorem to a paper affect its chance of acceptance? Does labeling a post with the author's gender affect the post popularity? This paper develops a method to estimate such causal effects from observational text data, adjusting for confounding features of the text such as the subject or writing quality. We assume that the text suffices for causal adjustment but that, in practice, it is prohibitively high-dimensional. To address this challenge, we develop causally sufficient embeddings, low-dimensional document representations that preserve sufficient information for causal identification and allow for efficient estimation of causal effects. Causally sufficient embeddings combine two ideas. The first is supervised dimensionality reduction: causal adjustment requires only the aspects of text that are predictive of both the treatment and outcome. The second is efficient language modeling: representations of text are designed to dispose of linguistically irrelevant information, and this information is also causally irrelevant. Our method adapts language models (specifically, word embeddings and topic models) to learn document embeddings that are able to predict both treatment and outcome. We study causally sufficient embeddings with semi-synthetic datasets and find that they improve causal estimation over related embedding methods. We illustrate the methods by answering the two motivating questions---the effect of a theorem on paper acceptance and the effect of a gender label on post popularity. Code and data available at https://github.com/vveitch/causal-text-embeddings-tf2}{github.com/vveitch/causal-text-embeddings-tf2
While data augmentation is an important trick to boost the accuracy of deep learning methods in computer vision tasks, its study in natural language tasks is still very limited. In this paper, we present a novel data augmentation method for neural machine translation. Different from previous augmentation methods that randomly drop, swap or replace words with other words in a sentence, we softly augment a randomly chosen word in a sentence by its contextual mixture of multiple related words. More accurately, we replace the one-hot representation of a word by a distribution (provided by a language model) over the vocabulary, i.e., replacing the embedding of this word by a weighted combination of multiple semantically similar words. Since the weights of those words depend on the contextual information of the word to be replaced, the newly generated sentences capture much richer information than previous augmentation methods. Experimental results on both small scale and large scale machine translation datasets demonstrate the superiority of our method over strong baselines.