Recently, there has been a growing interest in designing text generation systems from a discourse coherence perspective, e.g., modeling the interdependence between sentences. Still, recent BERT-based evaluation metrics cannot recognize coherence and fail to punish incoherent elements in system outputs. In this work, we introduce DiscoScore, a parametrized discourse metric, which uses BERT to model discourse coherence from different perspectives, driven by Centering theory. Our experiments encompass 16 non-discourse and discourse metrics, including DiscoScore and popular coherence models, evaluated on summarization and document-level machine translation (MT). We find that (i) the majority of BERT-based metrics correlate much worse with human rated coherence than early discourse metrics, invented a decade ago; (ii) the recent state-of-the-art BARTScore is weak when operated at system level -- which is particularly problematic as systems are typically compared in this manner. DiscoScore, in contrast, achieves strong system-level correlation with human ratings, not only in coherence but also in factual consistency and other aspects, and surpasses BARTScore by over 10 correlation points on average. Further, aiming to understand DiscoScore, we provide justifications to the importance of discourse coherence for evaluation metrics, and explain the superiority of one variant over another. Our code is available at \url{//github.com/AIPHES/DiscoScore}.
相關內容
Teaching morals is one of the most important purposes of storytelling. An essential ability for understanding and writing moral stories is bridging story plots and implied morals. Its challenges mainly lie in: (1) grasping knowledge about abstract concepts in morals, (2) capturing inter-event discourse relations in stories, and (3) aligning value preferences of stories and morals concerning good or bad behavior. In this paper, we propose two understanding tasks and two generation tasks to assess these abilities of machines. We present STORAL, a new dataset of Chinese and English human-written moral stories. We show the difficulty of the proposed tasks by testing various models with automatic and manual evaluation on STORAL. Furthermore, we present a retrieval-augmented algorithm that effectively exploits related concepts or events in training sets as additional guidance to improve performance on these tasks.
The COVID-19 pandemic is accompanied by a massive "infodemic" that makes it hard to identify concise and credible information for COVID-19-related questions, like incubation time, infection rates, or the effectiveness of vaccines. As a novel solution, our paper is concerned with designing a question-answering system based on modern technologies from natural language processing to overcome information overload and misinformation in pandemic situations. To carry out our research, we followed a design science research approach and applied Ingwersen's cognitive model of information retrieval interaction to inform our design process from a socio-technical lens. On this basis, we derived prescriptive design knowledge in terms of design requirements and design principles, which we translated into the construction of a prototypical instantiation. Our implementation is based on the comprehensive CORD-19 dataset, and we demonstrate our artifact's usefulness by evaluating its answer quality based on a sample of COVID-19 questions labeled by biomedical experts.
Pretrained language models can be effectively stimulated by textual prompts or demonstrations, especially in low-data scenarios. Recent works have focused on automatically searching discrete or continuous prompts or optimized verbalizers, yet studies for the demonstration are still limited. Concretely, the demonstration examples are crucial for an excellent final performance of prompt-tuning. In this paper, we propose a novel pluggable, extensible, and efficient approach named contrastive demonstration tuning, which is free of demonstration sampling. Furthermore, the proposed approach can be: (i) Plugged to any previous prompt-tuning approaches; (ii) Extended to widespread classification tasks with a large number of categories. Experimental results on 16 datasets illustrate that our method integrated with previous approaches LM-BFF and P-tuning can yield better performance. Code is available in //github.com/zjunlp/PromptKG/tree/main/research/Demo-Tuning.
For multilingual sequence-to-sequence pretrained language models (multilingual Seq2Seq PLMs), e.g. mBART, the self-supervised pretraining task is trained on a wide range of monolingual languages, e.g. 25 languages from commoncrawl, while the downstream cross-lingual tasks generally progress on a bilingual language subset, e.g. English-German, making there exists the cross-lingual data discrepancy, namely \textit{domain discrepancy}, and cross-lingual learning objective discrepancy, namely \textit{task discrepancy}, between the pretrain and finetune stages. To bridge the above cross-lingual domain and task gaps, we extend the vanilla pretrain-finetune pipeline with extra code-switching restore task. Specifically, the first stage employs the self-supervised code-switching restore task as a pretext task, allowing the multilingual Seq2Seq PLM to acquire some in-domain alignment information. And for the second stage, we continuously fine-tune the model on labeled data normally. Experiments on a variety of cross-lingual NLG tasks, including 12 bilingual translation tasks, 36 zero-shot translation tasks, and cross-lingual summarization tasks show our model outperforms the strong baseline mBART consistently. Comprehensive analyses indicate our approach could narrow the cross-lingual sentence representation distance and improve low-frequency word translation with trivial computational cost.
Pre-trained models are widely used in the tasks of natural language processing nowadays. However, in the specific field of text simplification, the research on improving pre-trained models is still blank. In this work, we propose a continued pre-training method for text simplification. Specifically, we propose a new masked language modeling (MLM) mechanism, which does not randomly mask words but only masks simple words. The new mechanism can make the model learn to generate simple words. We use a small-scale simple text dataset for continued pre-training and employ two methods to identify simple words from the texts. We choose BERT, a representative pre-trained model, and continue pre-training it using our proposed method. Finally, we obtain SimpleBERT, which surpasses BERT in both lexical simplification and sentence simplification tasks and has achieved state-of-the-art results on multiple datasets. What's more, SimpleBERT can replace BERT in existing simplification models without modification.
Due to the success of pre-trained language models, versions of languages other than English have been released in recent years. This fact implies the need for resources to evaluate these models. In the case of Spanish, there are few ways to systematically assess the models' quality. In this paper, we narrow the gap by building two evaluation benchmarks. Inspired by previous work (Conneau and Kiela, 2018; Chen et al., 2019), we introduce Spanish SentEval and Spanish DiscoEval, aiming to assess the capabilities of stand-alone and discourse-aware sentence representations, respectively. Our benchmarks include considerable pre-existing and newly constructed datasets that address different tasks from various domains. In addition, we evaluate and analyze the most recent pre-trained Spanish language models to exhibit their capabilities and limitations. As an example, we discover that for the case of discourse evaluation tasks, mBERT, a language model trained on multiple languages, usually provides a richer latent representation than models trained only with documents in Spanish. We hope our contribution will motivate a fairer, more comparable, and less cumbersome way to evaluate future Spanish language models.
Online review systems are the primary means through which many businesses seek to build the brand and spread their messages. Prior research studying the effects of online reviews has been mainly focused on a single numerical cause, e.g., ratings or sentiment scores. We argue that such notions of causes entail three key limitations: they solely consider the effects of single numerical causes and ignore different effects of multiple aspects -- e.g., Food, Service -- embedded in the textual reviews; they assume the absence of hidden confounders in observational studies, e.g., consumers' personal preferences; and they overlook the indirect effects of numerical causes that can potentially cancel out the effect of textual reviews on business revenue. We thereby propose an alternative perspective to this single-cause-based effect estimation of online reviews: in the presence of hidden confounders, we consider multi-aspect textual reviews, particularly, their total effects on business revenue and direct effects with the numerical cause -- ratings -- being the mediator. We draw on recent advances in machine learning and causal inference to together estimate the hidden confounders and causal effects. We present empirical evaluations using real-world examples to discuss the importance and implications of differentiating the multi-aspect effects in strategizing business operations.
Common image-text joint understanding techniques presume that images and the associated text can universally be characterized by a single implicit model. However, co-occurring images and text can be related in qualitatively different ways, and explicitly modeling it could improve the performance of current joint understanding models. In this paper, we train a Cross-Modal Coherence Modelfor text-to-image retrieval task. Our analysis shows that models trained with image--text coherence relations can retrieve images originally paired with target text more often than coherence-agnostic models. We also show via human evaluation that images retrieved by the proposed coherence-aware model are preferred over a coherence-agnostic baseline by a huge margin. Our findings provide insights into the ways that different modalities communicate and the role of coherence relations in capturing commonsense inferences in text and imagery.
Recent work pre-training Transformers with self-supervised objectives on large text corpora has shown great success when fine-tuned on downstream NLP tasks including text summarization. However, pre-training objectives tailored for abstractive text summarization have not been explored. Furthermore there is a lack of systematic evaluation across diverse domains. In this work, we propose pre-training large Transformer-based encoder-decoder models on massive text corpora with a new self-supervised objective. In PEGASUS, important sentences are removed/masked from an input document and are generated together as one output sequence from the remaining sentences, similar to an extractive summary. We evaluated our best PEGASUS model on 12 downstream summarization tasks spanning news, science, stories, instructions, emails, patents, and legislative bills. Experiments demonstrate it achieves state-of-the-art performance on all 12 downstream datasets measured by ROUGE scores. Our model also shows surprising performance on low-resource summarization, surpassing previous state-of-the-art results on 6 datasets with only 1000 examples. Finally we validated our results using human evaluation and show that our model summaries achieve human performance on multiple datasets.
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, XLNet outperforms BERT on 20 tasks, often by a large margin, and achieves state-of-the-art results on 18 tasks including question answering, natural language inference, sentiment analysis, and document ranking.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191