ACL 2020 论文列表

Recently, there has been much interest in the question of whether deep natural language understanding (NLU) models exhibit systematicity, generalizing such that units like words make consistent contributions to the meaning of the sentences in which they appear. There is accumulating evidence that neural models do not learn systematically. We examine the notion of systematicity from a linguistic perspective, defining a set of probing tasks and a set of metrics to measure systematic behaviour. We also identify ways in which network architectures can generalize non-systematically, and discuss why such forms of generalization may be unsatisfying. As a case study, we perform a series of experiments in the setting of natural language inference (NLI). We provide evidence that current state-of-the-art NLU systems do not generalize systematically, despite overall high performance.

In recent years there has been a burgeoning interest in the use of computational methods to distinguish between elicited speech samples produced by patients with dementia, and those from healthy controls. The difference between perplexity estimates from two neural language models (LMs) - one trained on transcripts of speech produced by healthy participants and one trained on those with dementia - as a single feature for diagnostic classification of unseen transcripts has been shown to produce state-of-the-art performance. However, little is known about why this approach is effective, and on account of the lack of case/control matching in the most widely-used evaluation set of transcripts (DementiaBank), it is unclear if these approaches are truly diagnostic, or are sensitive to other variables. In this paper, we interrogate neural LMs trained on participants with and without dementia by using synthetic narratives previously developed to simulate progressive semantic dementia by manipulating lexical frequency. We find that perplexity of neural LMs is strongly and differentially associated with lexical frequency, and that using a mixture model resulting from interpolating control and dementia LMs improves upon the current state-of-the-art for models trained on transcript text exclusively.

The supervised training of high-capacity models on large datasets containing hundreds of thousands of document-summary pairs is critical to the recent success of deep learning techniques for abstractive summarization. Unfortunately, in most domains (other than news) such training data is not available and cannot be easily sourced. In this paper we enable the use of supervised learning for the setting where there are only documents available (e.g., product or business reviews) without ground truth summaries. We create a synthetic dataset from a corpus of user reviews by sampling a review, pretending it is a summary, and generating noisy versions thereof which we treat as pseudo-review input. We introduce several linguistically motivated noise generation functions and a summarization model which learns to denoise the input and generate the original review. At test time, the model accepts genuine reviews and generates a summary containing salient opinions, treating those that do not reach consensus as noise. Extensive automatic and human evaluation shows that our model brings substantial improvements over both abstractive and extractive baselines.

Most general-purpose extractive summarization models are trained on news articles, which are short and present all important information upfront. As a result, such models are biased on position and often perform a smart selection of sentences from the beginning of the document. When summarizing long narratives, which have complex structure and present information piecemeal, simple position heuristics are not sufficient. In this paper, we propose to explicitly incorporate the underlying structure of narratives into general unsupervised and supervised extractive summarization models. We formalize narrative structure in terms of key narrative events (turning points) and treat it as latent in order to summarize screenplays (i.e., extract an optimal sequence of scenes). Experimental results on the CSI corpus of TV screenplays, which we augment with scene-level summarization labels, show that latent turning points correlate with important aspects of a CSI episode and improve summarization performance over general extractive algorithms leading to more complete and diverse summaries.

It is well known that the standard likelihood training and approximate decoding objectives in neural text generation models lead to less human-like responses for open-ended tasks such as language modeling and story generation. In this paper we have analyzed limitations of these models for abstractive document summarization and found that these models are highly prone to hallucinate content that is unfaithful to the input document. We conducted a large scale human evaluation of several neural abstractive summarization systems to better understand the types of hallucinations they produce. Our human annotators found substantial amounts of hallucinated content in all model generated summaries. However, our analysis does show that pretrained models are better summarizers not only in terms of raw metrics, i.e., ROUGE, but also in generating faithful and factual summaries as evaluated by humans. Furthermore, we show that textual entailment measures better correlate with faithfulness than standard metrics, potentially leading the way to automatic evaluation metrics as well as training and decoding criteria.

Sequence-to-sequence (seq2seq) network is a well-established model for text summarization task. It can learn to produce readable content; however, it falls short in effectively identifying key regions of the source. In this paper, we approach the content selection problem for clinical abstractive summarization by augmenting salient ontological terms into the summarizer. Our experiments on two publicly available clinical data sets (107,372 reports of MIMIC-CXR, and 3,366 reports of OpenI) show that our model statistically significantly boosts state-of-the-art results in terms of ROUGE metrics (with improvements: 2.9% RG-1, 2.5% RG-2, 1.9% RG-L), in the healthcare domain where any range of improvement impacts patients’ welfare.

Non-autoregressive (NAR) neural machine translation is usually done via knowledge distillation from an autoregressive (AR) model. Under this framework, we leverage large monolingual corpora to improve the NAR model’s performance, with the goal of transferring the AR model’s generalization ability while preventing overfitting. On top of a strong NAR baseline, our experimental results on the WMT14 En-De and WMT16 En-Ro news translation tasks confirm that monolingual data augmentation consistently improves the performance of the NAR model to approach the teacher AR model’s performance, yields comparable or better results than the best non-iterative NAR methods in the literature and helps reduce overfitting in the training process.

Subword segmentation is widely used to address the open vocabulary problem in machine translation. The dominant approach to subword segmentation is Byte Pair Encoding (BPE), which keeps the most frequent words intact while splitting the rare ones into multiple tokens. While multiple segmentations are possible even with the same vocabulary, BPE splits words into unique sequences; this may prevent a model from better learning the compositionality of words and being robust to segmentation errors. So far, the only way to overcome this BPE imperfection, its deterministic nature, was to create another subword segmentation algorithm (Kudo, 2018). In contrast, we show that BPE itself incorporates the ability to produce multiple segmentations of the same word. We introduce BPE-dropout - simple and effective subword regularization method based on and compatible with conventional BPE. It stochastically corrupts the segmentation procedure of BPE, which leads to producing multiple segmentations within the same fixed BPE framework. Using BPE-dropout during training and the standard BPE during inference improves translation quality up to 2.3 BLEU compared to BPE and up to 0.9 BLEU compared to the previous subword regularization.

This paper introduces a new task of politeness transfer which involves converting non-polite sentences to polite sentences while preserving the meaning. We also provide a dataset of more than 1.39 instances automatically labeled for politeness to encourage benchmark evaluations on this new task. We design a tag and generate pipeline that identifies stylistic attributes and subsequently generates a sentence in the target style while preserving most of the source content. For politeness as well as five other transfer tasks, our model outperforms the state-of-the-art methods on automatic metrics for content preservation, with a comparable or better performance on style transfer accuracy. Additionally, our model surpasses existing methods on human evaluations for grammaticality, meaning preservation and transfer accuracy across all the six style transfer tasks. The data and code is located at https://github.com/tag-and-generate.

We formulate the novel task of automatically updating an existing natural language comment based on changes in the body of code it accompanies. We propose an approach that learns to correlate changes across two distinct language representations, to generate a sequence of edits that are applied to the existing comment to reflect the source code modifications. We train and evaluate our model using a dataset that we collected from commit histories of open-source software projects, with each example consisting of a concurrent update to a method and its corresponding comment. We compare our approach against multiple baselines using both automatic metrics and human evaluation. Results reflect the challenge of this task and that our model outperforms baselines with respect to making edits.

Abstract Meaning Representations (AMRs) are broad-coverage sentence-level semantic graphs. Existing approaches to generating text from AMR have focused on training sequence-to-sequence or graph-to-sequence models on AMR annotated data only. In this paper, we propose an alternative approach that combines a strong pre-trained language model with cycle consistency-based re-scoring. Despite the simplicity of the approach, our experimental results show these models outperform all previous techniques on the English LDC2017T10 dataset, including the recent use of transformer architectures. In addition to the standard evaluation metrics, we provide human evaluation experiments that further substantiate the strength of our approach.

To address the challenge of policy learning in open-domain multi-turn conversation, we propose to represent prior information about dialog transitions as a graph and learn a graph grounded dialog policy, aimed at fostering a more coherent and controllable dialog. To this end, we first construct a conversational graph (CG) from dialog corpora, in which there are vertices to represent “what to say” and “how to say”, and edges to represent natural transition between a message (the last utterance in a dialog context) and its response. We then present a novel CG grounded policy learning framework that conducts dialog flow planning by graph traversal, which learns to identify a what-vertex and a how-vertex from the CG at each turn to guide response generation. In this way, we effectively leverage the CG to facilitate policy learning as follows: (1) it enables more effective long-term reward design, (2) it provides high-quality candidate actions, and (3) it gives us more control over the policy. Results on two benchmark corpora demonstrate the effectiveness of this framework.

Many multi-domain neural machine translation (NMT) models achieve knowledge transfer by enforcing one encoder to learn shared embedding across domains. However, this design lacks adaptation to individual domains. To overcome this limitation, we propose a novel multi-domain NMT model using individual modules for each domain, on which we apply word-level, adaptive and layer-wise domain mixing. We first observe that words in a sentence are often related to multiple domains. Hence, we assume each word has a domain proportion, which indicates its domain preference. Then word representations are obtained by mixing their embedding in individual domains based on their domain proportions. We show this can be achieved by carefully designing multi-head dot-product attention modules for different domains, and eventually taking weighted averages of their parameters by word-level layer-wise domain proportions. Through this, we can achieve effective domain knowledge sharing and capture fine-grained domain-specific knowledge as well. Our experiments show that our proposed model outperforms existing ones in several NMT tasks.

Recent advancements in neural language modelling make it possible to rapidly generate vast amounts of human-sounding text. The capabilities of humans and automatic discriminators to detect machine-generated text have been a large source of research interest, but humans and machines rely on different cues to make their decisions. Here, we perform careful benchmarking and analysis of three popular sampling-based decoding strategies—top-k, nucleus sampling, and untruncated random sampling—and show that improvements in decoding methods have primarily optimized for fooling humans. This comes at the expense of introducing statistical abnormalities that make detection easy for automatic systems. We also show that though both human and automatic detector performance improve with longer excerpt length, even multi-sentence excerpts can fool expert human raters over 30% of the time. Our findings reveal the importance of using both human and automatic detectors to assess the humanness of text generation systems.

Natural language understanding (NLU) and natural language generation (NLG) are two fundamental and related tasks in building task-oriented dialogue systems with opposite objectives: NLU tackles the transformation from natural language to formal representations, whereas NLG does the reverse. A key to success in either task is parallel training data which is expensive to obtain at a large scale. In this work, we propose a generative model which couples NLU and NLG through a shared latent variable. This approach allows us to explore both spaces of natural language and formal representations, and facilitates information sharing through the latent space to eventually benefit NLU and NLG. Our model achieves state-of-the-art performance on two dialogue datasets with both flat and tree-structured formal representations. We also show that the model can be trained in a semi-supervised fashion by utilising unlabelled data to boost its performance.

Predicting reading time has been a subject of much previous work, focusing on how different words affect human processing, measured by reading time. However, previous work has dealt with a limited number of participants as well as word level only predictions (i.e. predicting the time to read a single word). We seek to extend these works by examining whether or not document level predictions are effective, given additional information such as subject matter, font characteristics, and readability metrics. We perform a novel experiment to examine how different features of text contribute to the time it takes to read, distributing and collecting data from over a thousand participants. We then employ a large number of machine learning methods to predict a user’s reading time. We find that despite extensive research showing that word level reading time can be most effectively predicted by neural networks, larger scale text can be easily and most accurately predicted by one factor, the number of words.

Suspense is a crucial ingredient of narrative fiction, engaging readers and making stories compelling. While there is a vast theoretical literature on suspense, it is computationally not well understood. We compare two ways for modelling suspense: surprise, a backward-looking measure of how unexpected the current state is given the story so far; and uncertainty reduction, a forward-looking measure of how unexpected the continuation of the story is. Both can be computed either directly over story representations or over their probability distributions. We propose a hierarchical language model that encodes stories and computes surprise and uncertainty reduction. Evaluating against short stories annotated with human suspense judgements, we find that uncertainty reduction over representations is the best predictor, resulting in near human accuracy. We also show that uncertainty reduction can be used to predict suspenseful events in movie synopses.

With the advent of powerful neural language models over the last few years, research attention has increasingly focused on what aspects of language they represent that make them so successful. Several testing methodologies have been developed to probe models’ syntactic representations. One popular method for determining a model’s ability to induce syntactic structure trains a model on strings generated according to a template then tests the model’s ability to distinguish such strings from superficially similar ones with different syntax. We illustrate a fundamental problem with this approach by reproducing positive results from a recent paper with two non-syntactic baseline language models: an n-gram model and an LSTM model trained on scrambled inputs.

Can artificial neural networks learn to represent inflectional morphology and generalize to new words as human speakers do? Kirov and Cotterell (2018) argue that the answer is yes: modern Encoder-Decoder (ED) architectures learn human-like behavior when inflecting English verbs, such as extending the regular past tense form /-(e)d/ to novel words. However, their work does not address the criticism raised by Marcus et al. (1995): that neural models may learn to extend not the regular, but the most frequent class — and thus fail on tasks like German number inflection, where infrequent suffixes like /-s/ can still be productively generalized. To investigate this question, we first collect a new dataset from German speakers (production and ratings of plural forms for novel nouns) that is designed to avoid sources of information unavailable to the ED model. The speaker data show high variability, and two suffixes evince ‘regular’ behavior, appearing more often with phonologically atypical inputs. Encoder-decoder models do generalize the most frequently produced plural class, but do not show human-like variability or ‘regular’ extension of these other plural markers. We conclude that modern neural models may still struggle with minority-class generalization.