Dataset distillation aims to compress a training dataset by creating a small number of informative synthetic samples such that neural networks trained on them perform as well as those trained on the original training dataset. Current text dataset distillation methods create each synthetic sample as a sequence of word embeddings instead of a text to apply gradient-based optimization; however, such embedding-level distilled datasets cannot be used for training other models whose word embedding weights are different from the model used for distillation. To address this issue, we propose a novel text dataset distillation approach, called Distilling dataset into Language Model (DiLM), which trains a language model to generate informative synthetic training samples as text data, instead of directly optimizing synthetic samples. We evaluated DiLM on various text classification datasets and showed that distilled synthetic datasets from DiLM outperform those from current coreset selection methods. DiLM achieved remarkable generalization performance in training different types of models and in-context learning of large language models. Our code will be available at https://github.com/arumaekawa/DiLM.

Extractive summarization can produce faithful summaries but often requires additional constraints such as a desired summary length. Traditional sentence compression models do not typically consider the constraints because of their restricted model abilities, which require model modifications for coping with them. To bridge this gap, we propose Instruction-based Compression (InstructCMP), an approach to the sentence compression task that can consider the length constraint through instructions by leveraging the zero-shot task-solving abilities of Large Language Models (LLMs). For this purpose, we created new evaluation datasets by transforming traditional sentence compression datasets into an instruction format. By using the datasets, we first reveal that the current LLMs still face challenges in accurately controlling the length for a compressed text. To address this issue, we propose an approach named length priming, that incorporates additional length information into the instructions without external resources. While the length priming effectively works in a zero-shot setting, a training dataset with the instructions would further improve the ability of length control. Thus, we additionally created a training dataset in an instruction format to fine-tune the model on it. Experimental results and analysis show that applying the length priming significantly improves performances of InstructCMP in both zero-shot and fine-tuning settings without the need of any model modifications.

Due to biases inherently present in data for pre-training, current pre-trained Large Language Models (LLMs) also ubiquitously manifest the same phenomena. Since the bias influences the output from the LLMs across various tasks, the widespread deployment of the LLMs is hampered. We propose a simple method that utilizes structured knowledge to alleviate this issue, aiming to reduce the bias embedded within the LLMs and ensuring they have an encompassing perspective when used in applications. Experimental results indicated that our method has good debiasing ability when applied to existing both autoregressive and masked language models. Additionally, it could ensure that the performances of LLMs on downstream tasks remain uncompromised.Our method outperforms state-of-the-art (SOTA) baselines in the debiasing ability. Importantly, our method obviates the need for training from scratch, thus offering enhanced scalability and cost-effectiveness.

For abstractive text summarization, laborious data annotation and time-consuming model training become two high walls, hindering its further progress. Active Learning, selecting a few informative instances for annotation and model training, sheds light on solving these issues. However, only few active learning-based studies focus on abstractive text summarization and suffer from low stability, effectiveness, and efficiency. To solve the problems, we propose a novel LLM-determined curriculum active learning framework. Firstly, we design a prompt to ask large language models to rate the difficulty of instances, which guides the model to train on from easier to harder instances. Secondly, we design a novel active learning strategy, i.e., Certainty Gain Maximization, enabling to select instances whose distribution aligns well with the overall distribution. Experiments show our method can improve stability, effectiveness, and efficiency of abstractive text summarization backbones.

This paper tackles a new task: discourse parsing for videos, inspired by text discourse parsing based on Rhetorical Structure Theory (RST). The task aims to construct an RST tree for a video to represent its storyline and illustrate the event relationships. We first construct a benchmark dataset by identifying events with their time spans, providing corresponding captions, and constructing RST trees with events as leaves. We then evaluate baseline approaches to video RST parsing: the ‘parsing after captioning’ framework and parsing via visual features. The results show that a parser using gold captions performed the best, while parsers relying on generated captions performed the worst; a parser using visual features provided intermediate performance. However, we observed that parsing via visual features could be improved by pre-training it with video captioning designed to produce a coherent video story. Furthermore, we demonstrated that RST trees obtained from videos contribute to multimodal summarization consisting of keyframes with texts.

Multi-modal machine translation (MMT) can reduce ambiguity and semantic distortion compared with traditional machine translation (MT) by utilizing auxiliary information such as images. However, current MMT methods face two primary challenges. The first is their underperformance compared to MT methods based on pre-trained models. The second is the inadequate exploitation and integration of the image modality within the model, primarily due to a lack of triplet training data. A mainstream approach is to introduce large amounts of parallel and monolingual data to train the text model and the visual model separately. However, incorporating extensive external data can result in data imbalance, which may introduce biases during training. Additionally, the collection and cleaning of such large datasets is labor-intensive. To overcome these challenges, we introduce a novel, low-cost, large language model-based data augmentation method called LAMBDA, which can enrich the original samples and expand the dataset without requiring external images and text. We propose a fine-grained image captioning module with a noise filter to hierarchically and accurately extract unexploited information from images. Additionally, we design two specific prompts to guide the GPT-3.5 model in generating enriched texts and the corresponding translations. The enriched samples contain diverse text and strong connections between text and images, leading to significant improvements for MMT baselines, with the highest being an increase of up to 3.83 BLEU score and 3.61 METEOR score.

Ad text generation is vital for automatic advertising in various fields through search engine advertising (SEA) to avoid the cost problem caused by laborious human efforts for creating ad texts. Even though ad creators create the landing page (LP) for advertising and we can expect its quality, conventional approaches with reinforcement learning (RL) mostly focus on advertising keywords rather than LP information. This work investigates and shows the effective usage of LP information as a reward in RL-based ad text generation through automatic and human evaluations. Our analysis of the actually generated ad text shows that LP information can be a crucial reward by appropriately scaling its value range to improve ad text generation performance.

Recently, decoder-only pre-trained large language models (LLMs), with several tens of billion parameters, have significantly impacted a wide range of natural language processing (NLP) tasks. While encoder-only or encoder-decoder pre-trained language models have already proved to be effective in discourse parsing, the extent to which LLMs can perform this task remains an open research question. Therefore, this paper explores how beneficial such LLMs are for Rhetorical Structure Theory (RST) discourse parsing. Here, the parsing process for both fundamental top-down and bottom-up strategies is converted into prompts, which LLMs can work with. We employ Llama 2 and fine-tune it with QLoRA, which has fewer parameters that can be tuned. Experimental results on three benchmark datasets, RST-DT, Instr-DT, and the GUM corpus, demonstrate that Llama 2 with 70 billion parameters in the bottom-up strategy obtained state-of-the-art (SOTA) results with significant differences. Furthermore, our parsers demonstrated generalizability when evaluated on RST-DT, showing that, in spite of being trained with the GUM corpus, it obtained similar performances to those of existing parsers trained with RST-DT.

Although fine-tuning a pre-trained model with a conventional approach has shown to be effective in various downstream tasks, previous work has used only backpropagation to fine-tune the model, which causes a massive amount of computational resources and time. We propose Extreme Fine-Tuning (EFT), a novel approach for fine-tuning a pre-trained model effectively and efficiently. EFT uses backpropagation for a brief fine-tuning and an iterative extreme learning machine for training a classifier. We applied EFT to four text classification datasets, MELD, IEMOCAP, IMDb, and AG News, and compared its performance with state-of-the-art (SOTA) approaches. The results indicate that EFT noticeably outperformed the other approaches in training-time measurement with comparable model performance. We will release our code at https://github.com/up-33/extreme-fine-tuning.

Dataset distillation aims to create a small dataset of informative synthetic samples to rapidly train neural networks that retain the performance of the original dataset. In this paper, we focus on constructing distilled few-shot datasets for natural language processing (NLP) tasks to fine-tune pre-trained transformers. Specifically, we propose to introduce attention labels, which can efficiently distill the knowledge from the original dataset and transfer it to the transformer models via attention probabilities. We evaluated our dataset distillation methods in four various NLP tasks and demonstrated that it is possible to create distilled few-shot datasets with the attention labels, yielding impressive performances for fine-tuning BERT. Specifically, in AGNews, a four-class news classification task, our distilled few-shot dataset achieved up to 93.2% accuracy, which is 98.5% performance of the original dataset even with only one sample per class and only one gradient step.

In a controllable text generation dataset, there exist unannotated attributes that could provide irrelevant learning signals to models that use it for training and thus degrade their performance. We propose focused prefix tuning (FPT) to mitigate the problem and to enable the control to focus on the desired attribute. Experimental results show that FPT can achieve better control accuracy and text fluency than baseline models in single-attribute control tasks. In multi-attribute control tasks, FPT achieves comparable control accuracy with the state-of-the-art approach while keeping the flexibility to control new attributes without retraining existing models.

Recently, we can obtain a practical abstractive document summarization model by fine-tuning a pre-trained language model (PLM). Since the pre-training for PLMs does not consider summarization-specific information such as the target summary length, there is a gap between the pre-training and fine-tuning for PLMs in summarization tasks. To fill the gap, we propose a method for enabling the model to understand the summarization-specific information by predicting the summary length in the encoder and generating a summary of the predicted length in the decoder in fine-tuning. Experimental results on the WikiHow, NYT, and CNN/DM datasets showed that our methods improve ROUGE scores from BART by generating summaries of appropriate lengths. Further, we observed about 3.0, 1,5, and 3.1 point improvements for ROUGE-1, -2, and -L, respectively, from GSum on the WikiHow dataset. Human evaluation results also showed that our methods improve the informativeness and conciseness of summaries.

Pre-trained seq2seq models have achieved state-of-the-art results in the grammatical error correction task. However, these models still suffer from a prediction bias due to their unidirectional decoding. Thus, we propose a bidirectional Transformer reranker (BTR), that re-estimates the probability of each candidate sentence generated by the pre-trained seq2seq model. The BTR preserves the seq2seq-style Transformer architecture but utilizes a BERT-style self-attention mechanism in the decoder to compute the probability of each target token by using masked language modeling to capture bidirectional representations from the target context. For guiding the reranking, the BTR adopts negative sampling in the objective function to minimize the unlikelihood. During inference, the BTR gives final results after comparing the reranked top-1 results with the original ones by an acceptance threshold. Experimental results show that, in reranking candidates from a pre-trained seq2seq model, T5-base, the BTR on top of T5-base could yield 65.47 and 71.27 F0.5 scores on the CoNLL-14 and BEA test sets, respectively, and yield 59.52 GLEU score on the JFLEG corpus, with improvements of 0.36, 0.76 and 0.48 points compared with the original T5-base. Furthermore, when reranking candidates from T5-large, the BTR on top of T5-base improved the original T5-large by 0.26 points on the BEA test set.

Hybrid Question-Answering (HQA), which targets reasoning over tables and passages linked from table cells, has witnessed significant research in recent years. A common challenge in HQA and other passage-table QA datasets is that it is generally unrealistic to iterate over all table rows, columns, and linked passages to retrieve evidence. Such a challenge made it difficult for previous studies to show their reasoning ability in retrieving answers. To bridge this gap, we propose a novel Table-alignment-based Cell-selection and Reasoning model (TACR) for hybrid text and table QA, evaluated on the HybridQA and WikiTableQuestions datasets. In evidence retrieval, we design a table-question-alignment enhanced cell-selection method to retrieve fine-grained evidence. In answer reasoning, we incorporate a QA module that treats the row containing selected cells as context. Experimental results over the HybridQA and WikiTableQuestions (WTQ) datasets show that TACR achieves state-of-the-art results on cell selection and outperforms fine-grained evidence retrieval baselines on HybridQA, while achieving competitive performance on WTQ. We also conducted a detailed analysis to demonstrate that being able to align questions to tables in the cell-selection stage can result in important gains from experiments of over 90% table row and column selection accuracy, meanwhile also improving output explainability.

Multimodal emotion recognition aims to recognize emotions for each utterance from multiple modalities, which has received increasing attention for its application in human-machine interaction. Current graph-based methods fail to simultaneously depict global contextual features and local diverse uni-modal features in a dialogue. Furthermore, with the number of graph layers increasing, they easily fall into over-smoothing. In this paper, we propose a method for joint modality fusion and graph contrastive learning for multimodal emotion recognition (Joyful), where multimodality fusion, contrastive learning, and emotion recognition are jointly optimized. Specifically, we first design a new multimodal fusion mechanism that can provide deep interaction and fusion between the global contextual and uni-modal specific features. Then, we introduce a graph contrastive learning framework with inter- and intra-view contrastive losses to learn more distinguishable representations for samples with different sentiments. Extensive experiments on three benchmark datasets indicate that Joyful achieved state-of-the-art (SOTA) performance compared with all baselines. Code is released on Github (https://anonymous.4open.science/r/MERC-7F88).

Continual learning aims to accumulate knowledge to solve new tasks without catastrophic forgetting for previously learned tasks. Research on continual learning has led to the development of generative replay, which prevents catastrophic forgetting by generating pseudo-samples for previous tasks and learning them together with new tasks. Inspired by the biological brain, we propose the hippocampal memory indexing to enhance the generative replay by controlling sample generation using compressed features of previous training samples. It enables the generation of a specific training sample from previous tasks, thus improving the balance and quality of generated replay samples. Experimental results indicate that our method effectively controls the sample generation and consistently outperforms the performance of current generative replay methods.

The purpose of feedback comment generation is to provide useful feedback comments for a wide range of errors in learners’ essays from a language learning perspective. Since it is difficult to obtain appropriate comments at a practical level with rule-based or retrieval- based methods, we explore neural-based gen- erative methods with pre-trained models. We further assume the effectiveness of consider- ing grammatical terms in generating feedback comments. Specifically, this paper proposes T5-based models using predicted grammati- cal terms, submitted to FCG GenChal, and presents their results. By using correct gram- matical terms, our model could improve the BLEU score by 19.0 points, compared with the baseline T5 without grammatical terms on the development dataset. Furthermore, by using predicted grammatical terms, our model could improve the manual evaluation score by 2.33 points, compared with the baseline T5 without grammatical terms on the test dataset.

The emergence of pre-trained language models has taken story generation, which is the task of automatically generating a comprehensible story from limited information, to a new stage. Although generated stories from the language models are fluent and grammatically correct, the lack of coherence affects their quality. We propose a knowledge-based multi-stage model that incorporates the schema, a kind of structured knowledge, to guide coherent story generation. Our framework includes a schema acquisition module, a plot generation module, and a surface realization module. In the schema acquisition module, high-relevant structured knowledge pieces are selected as a schema. In the plot generation module, a coherent plot plan is navigated by the schema. In the surface realization module, conditioned by the generated plot, a story is generated. Evaluations show that our methods can generate more comprehensible stories than strong baselines, especially with higher global coherence and less repetition.

This paper proposes a new second language learning task of generating a response including specified grammatical items. We consider two approaches: 1) fine-tuning a pre-trained language model (DialoGPT) by reinforcement learning and 2) providing a few-shot prompt to a large language model (GPT-3). For reinforcement learning, we examine combinations of three reward functions that consider grammatical items, diversity, and fluency. Our experiments confirm that both approaches can generate responses including the specified grammatical items and that it is crucial to consider fluency rather than diversity as the reward function.

A repetition is a response that repeats words in the previous speaker’s utterance in a dialogue. Repetitions are essential in communication to build trust with others, as investigated in linguistic studies. In this work, we focus on repetition generation. To the best of our knowledge, this is the first neural approach to address repetition generation. We propose Weighted Label Smoothing, a smoothing method for explicitly learning which words to repeat during fine-tuning, and a repetition scoring method that can output more appropriate repetitions during decoding. We conducted automatic and human evaluations involving applying these methods to the pre-trained language model T5 for generating repetitions. The experimental results indicate that our methods outperformed baselines in both evaluations.

Previous studies on the timeline summarization (TLS) task ignored the information interaction between sentences and dates, and adopted pre-defined unlearnable representations for them. They also considered date selection and event detection as two independent tasks, which makes it impossible to integrate their advantages and obtain a globally optimal summary. In this paper, we present a joint learning-based heterogeneous graph attention network for TLS (HeterTls), in which date selection and event detection are combined into a unified framework to improve the extraction accuracy and remove redundant sentences simultaneously. Our heterogeneous graph involves multiple types of nodes, the representations of which are iteratively learned across the heterogeneous graph attention layer. We evaluated our model on four datasets, and found that it significantly outperformed the current state-of-the-art baselines with regard to ROUGE scores and date selection metrics.

Writing an ad text that attracts people and persuades them to click or act is essential for the success of search engine advertising. Therefore, ad creators must consider various aspects of advertising appeals (A³) such as the price, product features, and quality. However, products and services exhibit unique effective A³ for different industries. In this work, we focus on exploring the effective A³ for different industries with the aim of assisting the ad creation process. To this end, we created a dataset of advertising appeals and used an existing model that detects various aspects for ad texts. Our experiments demonstrated %through correlation analysis that different industries have their own effective A³ and that the identification of the A³ contributes to the estimation of advertising performance.

Recent work has shown advantages of incorporating knowledge graphs (KGs) into BERT for various NLP tasks. One common way is to feed entity embeddings as an additional input during pre-training. There are two limitations to such a method. First, to train the entity embeddings to include rich information of factual knowledge, it typically requires access to the entire KG. This is challenging for KGs with daily changes (e.g., Wikidata). Second, it requires a large scale pre-training corpus with entity annotations and high computational cost during pre-training. In this work, we efficiently construct entity embeddings only from the type knowledge, that does not require access to the entire KG. Although the entity embeddings contain only local information, they perform very well when combined with context. Furthermore, we show that our entity embeddings, constructed from BERT’s input embeddings, can be directly incorporated into the fine-tuning phase without requiring any specialized pre-training. In addition, these entity embeddings can also be constructed on the fly without requiring a large memory footprint to store them. Finally, we propose task-specific models that incorporate our entity embeddings for entity linking, entity typing, and relation classification. Experiments show that our models have comparable or superior performance to existing models while being more resource efficient.

To promote and further develop RST-style discourse parsing models, we need a strong baseline that can be regarded as a reference for reporting reliable experimental results. This paper explores a strong baseline by integrating existing simple parsing strategies, top-down and bottom-up, with various transformer-based pre-trained language models.The experimental results obtained from two benchmark datasets demonstrate that the parsing performance strongly relies on the pre-trained language models rather than the parsing strategies.In particular, the bottom-up parser achieves large performance gains compared to the current best parser when employing DeBERTa.We further reveal that language models with a span-masking scheme especially boost the parsing performance through our analysis within intra- and multi-sentential parsing, and nuclearity prediction.

Text infilling aims to restore incomplete texts by filling in blanks, which has attracted more attention recently because of its wide application in ancient text restoration and text rewriting. However, attribute- aware text infilling is yet to be explored, and existing methods seldom focus on the infilling length of each blank or the number/location of blanks. In this paper, we propose an Attribute-aware Text Infilling method via a Pre-trained language model (A-TIP), which contains a text infilling component and a plug- and-play discriminator. Specifically, we first design a unified text infilling component with modified attention mechanisms and intra- and inter-blank positional encoding to better perceive the number of blanks and the infilling length for each blank. Then, we propose a plug-and-play discriminator to guide generation towards the direction of improving attribute relevance without decreasing text fluency. Finally, automatic and human evaluations on three open-source datasets indicate that A-TIP achieves state-of- the-art performance compared with all baselines.

Automated radiology report generation aims to generate paragraphs that describe fine-grained visual differences among cases, especially those between the normal and the diseased. Existing methods seldom consider the cross-modal alignment between textual and visual features and tend to ignore disease tags as an auxiliary for report generation. To bridge the gap between textual and visual information, in this study, we propose a “Jointly learning framework for automated disease Prediction and radiology report Generation (JPG)” to improve the quality of reports through the interaction between the main task (report generation) and two auxiliary tasks (feature alignment and disease prediction). The feature alignment and disease prediction help the model learn text-correlated visual features and record diseases as keywords so that it can output high-quality reports. Besides, the improved reports in turn provide additional harder samples for feature alignment and disease prediction to learn more precise visual and textual representations and improve prediction accuracy. All components are jointly trained in a manner that helps improve them iteratively and progressively. Experimental results demonstrate the effectiveness of JPG on the most commonly used IU X-RAY dataset, showing its superior performance over multiple state-of-the-art image captioning and medical report generation methods with regard to BLEU, METEOR, and ROUGE metrics.

Target-oriented Opinion Words Extraction (TOWE) is a fine-grained sentiment analysis task that aims to extract the corresponding opinion words of a given opinion target from the sentence. Recently, deep learning approaches have made remarkable progress on this task. Nevertheless, the TOWE task still suffers from the scarcity of training data due to the expensive data annotation process. Limited labeled data increase the risk of distribution shift between test data and training data. In this paper, we propose exploiting massive unlabeled data to reduce the risk by increasing the exposure of the model to varying distribution shifts. Specifically, we propose a novel Multi-Grained Consistency Regularization (MGCR) method to make use of unlabeled data and design two filters specifically for TOWE to filter noisy data at different granularity. Extensive experimental results on four TOWE benchmark datasets indicate the superiority of MGCR compared with current state-of-the-art methods. The in-depth analysis also demonstrates the effectiveness of the different-granularity filters.

Recent neural text generation models have shown significant improvement in generating descriptive text from structured data such as table formats. One of the remaining important challenges is generating more analytical descriptions that can be inferred from facts in a data source. The use of a template-based generator and a pointer-generator is among the potential alternatives for table-to-text generators. In this paper, we propose a framework consisting of a pre-trained model and a copy mechanism. The pre-trained models are fine-tuned to produce fluent text that is enriched with numerical reasoning. However, it still lacks fidelity to the table contents. The copy mechanism is incorporated in the fine-tuning step by using general placeholders to avoid producing hallucinated phrases that are not supported by a table while preserving high fluency. In summary, our contributions are (1) a new dataset for numerical table-to-text generation using pairs of a table and a paragraph of a table description with richer inference from scientific papers, and (2) a table-to-text generation framework enriched with numerical reasoning.

Character-based word-segmentation models have been extensively applied to agglutinative languages, including Thai, due to their high performance. These models estimate word boundaries from a character sequence. However, a character unit in sequences has no essential meaning, compared with word, subword, and character cluster units. We propose a Thai word-segmentation model that uses various types of information, including words, subwords, and character clusters, from a character sequence. Our model applies multiple attentions to refine segmentation inferences by estimating the significant relationships among characters and various unit types. The experimental results indicate that our model can outperform other state-of-the-art Thai word-segmentation models.

Character-aware neural language models can capture the relationship between words by exploiting character-level information and are particularly effective for languages with rich morphology. However, these models are usually biased towards information from surface forms. To alleviate this problem, we propose a simple and effective method to improve a character-aware neural language model by forcing a character encoder to produce word-based embeddings under Skip-gram architecture in a warm-up step without extra training data. We empirically show that the resulting character-aware neural language model achieves obvious improvements of perplexity scores on typologically diverse languages, that contain many low-frequency or unseen words.

In the social media, users frequently use small images called emojis in their posts. Although using emojis in texts plays a key role in recent communication systems, less attention has been paid on their positions in the given texts, despite that users carefully choose and put an emoji that matches their post. Exploring positions of emojis in texts will enhance understanding of the relationship between emojis and texts. We extend an emoji label prediction task taking into account the information of emoji positions, by jointly learning the emoji position in a tweet to predict the emoji label. The results demonstrate that the position of emojis in texts is a good clue to boost the performance of emoji label prediction. Human evaluation validates that there exists a suitable emoji position in a tweet, and our proposed task is able to make tweets more fancy and natural. In addition, considering emoji position can further improve the performance for the irony detection task compared to the emoji label prediction. We also report the experimental results for the modified dataset, due to the problem of the original dataset for the first shared task to predict an emoji label in SemEval2018.

Neural sequence-to-sequence (Seq2Seq) models and BERT have achieved substantial improvements in abstractive document summarization (ADS) without and with pre-training, respectively. However, they sometimes repeatedly attend to unimportant source phrases while mistakenly ignore important ones. We present reconstruction mechanisms on two levels to alleviate this issue. The sequence-level reconstructor reconstructs the whole document from the hidden layer of the target summary, while the word embedding-level one rebuilds the average of word embeddings of the source at the target side to guarantee that as much critical information is included in the summary as possible. Based on the assumption that inverse document frequency (IDF) measures how important a word is, we further leverage the IDF weights in our embedding-level reconstructor. The proposed frameworks lead to promising improvements for ROUGE metrics and human rating on both the CNN/Daily Mail and Newsroom summarization datasets.

Encoder-decoder models have been commonly used for many tasks such as machine translation and response generation. As previous research reported, these models suffer from generating redundant repetition. In this research, we propose a new mechanism for encoder-decoder models that estimates the semantic difference of a source sentence before and after being fed into the encoder-decoder model to capture the consistency between two sides. This mechanism helps reduce repeatedly generated tokens for a variety of tasks. Evaluation results on publicly available machine translation and response generation datasets demonstrate the effectiveness of our proposal.

Most of the previous Rhetorical Structure Theory (RST) parsing methods are based on supervised learning such as neural networks, that require an annotated corpus of sufficient size and quality. However, the RST Discourse Treebank (RST-DT), the benchmark corpus for RST parsing in English, is small due to the costly annotation of RST trees. The lack of large annotated training data causes poor performance especially in relation labeling. Therefore, we propose a method for improving neural RST parsing models by exploiting silver data, i.e., automatically annotated data. We create large-scale silver data from an unlabeled corpus by using a state-of-the-art RST parser. To obtain high-quality silver data, we extract agreement subtrees from RST trees for documents built using the RST parsers. We then pre-train a neural RST parser with the obtained silver data and fine-tune it on the RST-DT. Experimental results show that our method achieved the best micro-F1 scores for Nuclearity and Relation at 75.0 and 63.2, respectively. Furthermore, we obtained a remarkable gain in the Relation score, 3.0 points, against the previous state-of-the-art parser.

Although there are many studies on neural language generation (NLG), few trials are put into the real world, especially in the advertising domain. Generating ads with NLG models can help copywriters in their creation. However, few studies have adequately evaluated the effect of generated ads with actual serving included because it requires a large amount of training data and a particular environment. In this paper, we demonstrate a practical use case of generating ad-text with an NLG model. Specially, we show how to improve the ads’ impact, deploy models to a product, and evaluate the generated ads.

The task of generating weather-forecast comments from meteorological simulations has the following requirements: (i) the changes in numerical values for various physical quantities need to be considered, (ii) the weather comments should be dependent on delivery time and area information, and (iii) the comments should provide useful information for users. To meet these requirements, we propose a data-to-text model that incorporates three types of encoders for numerical forecast maps, observation data, and meta-data. We also introduce weather labels representing weather information, such as sunny and rain, for our model to explicitly describe useful information. We conducted automatic and human evaluations. The results indicate that our model performed best against baselines in terms of informativeness. We make our code and data publicly available.

Numerical tables are widely used to present experimental results in scientific papers. For table understanding, a metric-type is essential to discriminate numbers in the tables. We introduce a new information extraction task, metric-type identification from multi-level header numerical tables, and provide a dataset extracted from scientific papers consisting of header tables, captions, and metric-types. We then propose two joint-learning neural classification and generation schemes featuring pointer-generator-based and BERT-based models. Our results show that the joint models can handle both in-header and out-of-header metric-type identification problems.

This work presents multi-modal deep SVDD (mSVDD) for one-class text classification. By extending the uni-modal SVDD to a multiple modal one, we build mSVDD with multiple hyperspheres, that enable us to build a much better description for target one-class data. Additionally, the end-to-end architecture of mSVDD can jointly handle neural feature learning and one-class text learning. We also introduce a mechanism for incorporating negative supervision in the absence of real negative data, which can be beneficial to the mSVDD model. We conduct experiments on Reuters and 20 Newsgroup datasets, and the experimental results demonstrate that mSVDD outperforms uni-modal SVDD and mSVDD can get further improvements when negative supervision is incorporated.

One way to enhance user engagement in search engines is to suggest interesting facts to the user. Although relationships between persons are important as a target for text mining, there are few effective approaches for extracting the interesting relationships between persons. We therefore propose a method for extracting interesting relationships between persons from natural language texts by focusing on their surprisingness. Our method first extracts all personal relationships from dependency trees for the texts and then calculates surprise scores for distributed representations of the extracted relationships in an unsupervised manner. The unique point of our method is that it does not require any labeled dataset with annotation for the surprising personal relationships. The results of the human evaluation show that the proposed method could extract more interesting relationships between persons from Japanese Wikipedia articles than a popularity-based baseline method. We demonstrate our proposed method as a chrome plugin on google search.

Discourse segmentation and sentence-level discourse parsing play important roles for various NLP tasks to consider textual coherence. Despite recent achievements in both tasks, there is still room for improvement due to the scarcity of labeled data. To solve the problem, we propose a language model-based generative classifier (LMGC) for using more information from labels by treating the labels as an input while enhancing label representations by embedding descriptions for each label. Moreover, since this enables LMGC to make ready the representations for labels, unseen in the pre-training step, we can effectively use a pre-trained language model in LMGC. Experimental results on the RST-DT dataset show that our LMGC achieved the state-of-the-art F1 score of 96.72 in discourse segmentation. It further achieved the state-of-the-art relation F1 scores of 84.69 with gold EDU boundaries and 81.18 with automatically segmented boundaries, respectively, in sentence-level discourse parsing.

Sentence extractive summarization shortens a document by selecting sentences for a summary while preserving its important contents. However, constructing a coherent and informative summary is difficult using a pre-trained BERT-based encoder since it is not explicitly trained for representing the information of sentences in a document. We propose a nested tree-based extractive summarization model on RoBERTa (NeRoBERTa), where nested tree structures consist of syntactic and discourse trees in a given document. Experimental results on the CNN/DailyMail dataset showed that NeRoBERTa outperforms baseline models in ROUGE. Human evaluation results also showed that NeRoBERTa achieves significantly better scores than the baselines in terms of coherence and yields comparable scores to the state-of-the-art models.

Ranking the user comments posted on a news article is important for online news services because comment visibility directly affects the user experience. Research on ranking comments with different metrics to measure the comment quality has shown “constructiveness” used in argument analysis is promising from a practical standpoint. In this paper, we report a case study in which this constructiveness is examined in the real world. Specifically, we examine an in-house competition to improve the performance of ranking constructive comments and demonstrate the effectiveness of the best obtained model for a commercial service.

We tackle the task of automatically generating a function name from source code. Existing generators face difficulties in generating low-frequency or out-of-vocabulary subwords. In this paper, we propose two strategies for copying low-frequency or out-of-vocabulary subwords in inputs. Our best performing model showed an improvement over the conventional method in terms of our modified F1 and accuracy on the Java-small and Java-large datasets.

We propose neural models that can normalize text by considering the similarities of word strings and sounds. We experimentally compared a model that considers the similarities of both word strings and sounds, a model that considers only the similarity of word strings or of sounds, and a model without the similarities as a baseline. Results showed that leveraging the word string similarity succeeded in dealing with misspellings and abbreviations, and taking into account the sound similarity succeeded in dealing with phonetic substitutions and emphasized characters. So that the proposed models achieved higher F1 scores than the baseline.

Recently, automatic trivia fact extraction has attracted much research interest. Modern search engines have begun to provide trivia facts as the information for entities because they can motivate more user engagement. In this paper, we propose a new unsupervised algorithm that automatically mines trivia facts for a given entity. Unlike previous studies, the proposed algorithm targets at a single Wikipedia article and leverages its hierarchical structure via top-down processing. Thus, the proposed algorithm offers two distinctive advantages: it does not incur high computation time, and it provides a domain-independent approach for extracting trivia facts. Experimental results demonstrate that the proposed algorithm is over 100 times faster than the existing method which considers Wikipedia categories. Human evaluation demonstrates that the proposed algorithm can mine better trivia facts regardless of the target entity domain and outperforms the existing methods.

In community-based question answering (CQA) platforms, it takes time for a user to get useful information from among many answers. Although one solution is an answer ranking method, the user still needs to read through the top-ranked answers carefully. This paper proposes a new task of selecting a diverse and non-redundant answer set rather than ranking the answers. Our method is based on determinantal point processes (DPPs), and it calculates the answer importance and similarity between answers by using BERT. We built a dataset focusing on a Japanese CQA site, and the experiments on this dataset demonstrated that the proposed method outperformed several baseline methods.

We propose a simple and effective method for incorporating word clusters into the Continuous Bag-of-Words (CBOW) model. Specifically, we propose to replace infrequent input and output words in CBOW model with their clusters. The resulting cluster-incorporated CBOW model produces embeddings of frequent words and a small amount of cluster embeddings, which will be fine-tuned in downstream tasks. We empirically show our replacing method works well on several downstream tasks. Through our analysis, we show that our method might be also useful for other similar models which produce word embeddings.

We propose a global optimization method under length constraint (GOLC) for neural text summarization models. GOLC increases the probabilities of generating summaries that have high evaluation scores, ROUGE in this paper, within a desired length. We compared GOLC with two optimization methods, a maximum log-likelihood and a minimum risk training, on CNN/Daily Mail and a Japanese single document summarization data set of The Mainichi Shimbun Newspapers. The experimental results show that a state-of-the-art neural summarization model optimized with GOLC generates fewer overlength summaries while maintaining the fastest processing speed; only 6.70% overlength summaries on CNN/Daily and 7.8% on long summary of Mainichi, compared to the approximately 20% to 50% on CNN/Daily Mail and 10% to 30% on Mainichi with the other optimization methods. We also demonstrate the importance of the generation of in-length summaries for post-editing with the dataset Mainich that is created with strict length constraints. The ex- perimental results show approximately 30% to 40% improved post-editing time by use of in-length summaries.

Ranking comments on an online news service is a practically important task for the service provider, and thus there have been many studies on this task. However, most of them considered users’ positive feedback, such as “Like”-button clicks, as a quality measure. In this paper, we address directly evaluating the quality of comments on the basis of “constructiveness,” separately from user feedback. To this end, we create a new dataset including 100K+ Japanese comments with constructiveness scores (C-scores). Our experiments clarify that C-scores are not always related to users’ positive feedback, and the performance of pairwise ranking models tends to be enhanced by the variation of comments rather than articles.

We propose a simple and effective method to inject word-level information into character-aware neural language models. Unlike previous approaches which usually inject word-level information at the input of a long short-term memory (LSTM) network, we inject it into the softmax function. The resultant model can be seen as a combination of character-aware language model and simple word-level language model. Our injection method can also be used together with previous methods. Through the experiments on 14 typologically diverse languages, we empirically show that our injection method, when used together with the previous methods, works better than the previous methods, including a gating mechanism, averaging, and concatenation of word vectors. We also provide a comprehensive comparison of these injection methods.

Rhetorical Structure Theory (RST) parsing is crucial for many downstream NLP tasks that require a discourse structure for a text. Most of the previous RST parsers have been based on supervised learning approaches. That is, they require an annotated corpus of sufficient size and quality, and heavily rely on the language and domain dependent corpus. In this paper, we present two language-independent unsupervised RST parsing methods based on dynamic programming. The first one builds the optimal tree in terms of a dissimilarity score function that is defined for splitting a text span into smaller ones. The second builds the optimal tree in terms of a similarity score function that is defined for merging two adjacent spans into a large one. Experimental results on English and German RST treebanks showed that our parser based on span merging achieved the best score, around 0.8 F₁ score, which is close to the scores of the previous supervised parsers.

We present neural machine translation models for translating a sentence in a text by using a graph-based encoder which can consider coreference relations provided within the text explicitly. The graph-based encoder can dynamically encode the source text without attending to all tokens in the text. In experiments, our proposed models provide statistically significant improvement to the previous approach of at most 0.9 points in the BLEU score on the OpenSubtitle2018 English-to-Japanese data set. Experimental results also show that the graph-based encoder can handle a longer text well, compared with the previous approach.

Browsing news articles on multiple devices is now possible. The lengths of news article headlines have precise upper bounds, dictated by the size of the display of the relevant device or interface. Therefore, controlling the length of headlines is essential when applying the task of headline generation to news production. However, because there is no corpus of headlines of multiple lengths for a given article, previous research on controlling output length in headline generation has not discussed whether the system outputs could be adequately evaluated without multiple references of different lengths. In this paper, we introduce two corpora, which are Japanese News Corpus (JNC) and JApanese MUlti-Length Headline Corpus (JAMUL), to confirm the validity of previous evaluation settings. The JNC provides common supervision data for headline generation. The JAMUL is a large-scale evaluation dataset for headlines of three different lengths composed by professional editors. We report new findings on these corpora; for example, although the longest length reference summary can appropriately evaluate the existing methods controlling output length, this evaluation setting has several problems.

Discourse relations between sentences are often represented as a tree, and the tree structure provides important information for summarizers to create a short and coherent summary. However, current neural network-based summarizers treat the source document as just a sequence of sentences and ignore the tree-like discourse structure inherent in the document. To incorporate the information of a discourse tree structure into the neural network-based summarizers, we propose a discourse-aware neural extractive summarizer which can explicitly take into account the discourse dependency tree structure of the source document. Our discourse-aware summarizer can jointly learn the discourse structure and the salience score of a sentence by using novel hierarchical attention modules, which can be trained on automatically parsed discourse dependency trees. Experimental results showed that our model achieved competitive or better performances against state-of-the-art models in terms of ROUGE scores on the DailyMail dataset. We further conducted manual evaluations. The results showed that our approach also gained the coherence of the output summaries.

This study proposes a new neural machine translation (NMT) model based on the encoder-decoder model that incorporates named entity (NE) tags of source-language sentences. Conventional NMT models have two problems enumerated as follows: (i) they tend to have difficulty in translating words with multiple meanings because of the high ambiguity, and (ii) these models’abilitytotranslatecompoundwordsseemschallengingbecausetheencoderreceivesaword, a part of the compound word, at each time step. To alleviate these problems, the encoder of the proposed model encodes the input word on the basis of its NE tag at each time step, which could reduce the ambiguity of the input word. Furthermore,the encoder introduces a chunk-level LSTM layer over a word-level LSTM layer and hierarchically encodes a source-language sentence to capture a compound NE as a chunk on the basis of the NE tags. We evaluate the proposed model on an English-to-Japanese translation task with the ASPEC, and English-to-Bulgarian and English-to-Romanian translation tasks with the Europarl corpus. The evaluation results show that the proposed model achieves up to 3.11 point improvement in BLEU.

Recent neural models for response generation show good results in terms of general responses. In real conversations, however, depending on the speaker/responder, similar utterances should require different responses. In this study, we attempt to consider individual user’s information in adjusting the notable sequence-to-sequence (seq2seq) model for more diverse, user-specific responses. We assume that we need user-specific features to adjust the response and we argue that some selected representative words from the users are suitable for this task. Furthermore, we prove that even for unseen or unknown users, our model can provide more diverse and interesting responses, while maintaining correlation with input utterances. Experimental results with human evaluation show that our model can generate more interesting responses than the popular seq2seqmodel and achieve higher relevance with input utterances than our baseline.

In this research, we propose the task of question summarization. We first analyzed question-summary pairs extracted from a Community Question Answering (CQA) site, and found that a proportion of questions cannot be summarized by extractive approaches but requires abstractive approaches. We created a dataset by regarding the question-title pairs posted on the CQA site as question-summary pairs. By using the data, we trained extractive and abstractive summarization models, and compared them based on ROUGE scores and manual evaluations. Our experimental results show an abstractive method using an encoder-decoder model with a copying mechanism achieves better scores for both ROUGE-2 F-measure and the evaluations by human judges.

The sequence-to-sequence (Seq2Seq) model has been successfully applied to machine translation (MT). Recently, MT performances were improved by incorporating supervised attention into the model. In this paper, we introduce supervised attention to constituency parsing that can be regarded as another translation task. Evaluation results on the PTB corpus showed that the bracketing F-measure was improved by supervised attention.

We focus on non-standard usages of common words on social media. In the context of social media, words sometimes have other usages that are totally different from their original. In this study, we attempt to distinguish non-standard usages on social media from standard ones in an unsupervised manner. Our basic idea is that non-standardness can be measured by the inconsistency between the expected meaning of the target word and the given context. For this purpose, we use context embeddings derived from word embeddings. Our experimental results show that the model leveraging the context embedding outperforms other methods and provide us with findings, for example, on how to construct context embeddings and which corpus to use.

In English, high-quality sentence compression models by deleting words have been trained on automatically created large training datasets. We work on Japanese sentence compression by a similar approach. To create a large Japanese training dataset, a method of creating English training dataset is modified based on the characteristics of the Japanese language. The created dataset is used to train Japanese sentence compression models based on the recurrent neural network.

This paper describes automatic terminology intelligibility estimation for readership-oriented technical writing. We assume that the term frequency weighted by the types of documents can be an indicator of the term intelligibility for a certain readership. From this standpoint, we analyzed the relationship between the following: average intelligibility levels of 46 technical terms that were rated by about 120 laymen; numbers of documents that an Internet search