Automatic Readability Assessment (ARA) predicts the level of difficulty of a text, e.g. at Grade 1 to Grade 12. ARA is an ordinal classification task since the predicted levels follow an underlying order, from easy to difficult. However, most neural ARA models ignore the distance between the gold level and predicted level, treating all levels as independent labels. This paper investigates whether distance-sensitive loss functions can improve ARA performance. We evaluate a variety of loss functions on neural ARA models, and show that ordinal log-loss can produce statistically significant improvement over the standard cross-entropy loss in terms of adjacent accuracy in a majority of our datasets.
Sentence Simplification aims to make sentences easier to read and understand. With most effort on corpus development focused on English, the amount of annotated data is limited in Chinese. To address this need, we introduce CSSWiki, an open-source dataset for Chinese sentence simplification based on Wikipedia. This dataset contains 1.6k source sentences paired with their simplified versions. Each sentence pair is annotated with operation tags that distinguish between linguistic and content modifications. We analyze differences in annotation scheme and data statistics between CSSWiki and existing datasets. We then report baseline sentence simplification performance on CSSWiki using zero-shot and few-shot approaches with Large Language Models.
As technical fields become ever more specialized, and with continuous emergence of novel technical terms, it may not be always possible to avail of bilingual experts in the field to perform translation. This paper investigates the performance of bilingual non-experts in Computer-Assisted Translation. The translators were asked to identify and correct errors in MT output of technical terms in patent materials, aided only by example bilingual sentences. Targeting English-to-Chinese translation, we automatically extract the example sentences from a bilingual corpus of English and Chinese patents. We identify the most frequent translation candidates of a term, and then select the most relevant example sentences for each candidate according to semantic similarity. Even when given only two example sentences for each translation candidate, the non-expert translators were able to post-edit effectively, correcting 67.2% of the MT errors while mistakenly revising correct MT output in only 17% of the cases.
The recent introduction of ChatGPT has caused much stir in the translation industry because of its impressive translation performance against leaders in the industry. We review some ma-jor issues based on the BLEU comparisons of Chinese-to-English (C2E) and English-to-Chinese (E2C) machine translation (MT) performance by ChatGPT against a range of leading MT providers in mostly technical domains. Based on sample aligned sentences from a sizable bilingual Chinese-English patent corpus and other sources, we find that while ChatGPT perform better generally, it does not consistently perform better than others in all areas or cases. We also draw on novice translators as post-editors to explore a major component in MT post-editing: Optimization of terminology. Many new technical words, including MWEs (Multi-Word Expressions), are problematic because they involve terminological developments which must balance between proper encapsulation of technical innovation and conforming to past traditions . Drawing on the above-mentioned corpus we have been developing an AI mediated MT post-editing (MTPE) system through the optimization of precedent rendition distribution and semantic association to enhance the work of translators and MTPE practitioners.
The importance of multiword expressions (MWEs) for language learning is well established. While MWE research has been evaluated on various downstream tasks such as syntactic parsing and machine translation, its applications in computer-assisted language learning has been less explored. This paper investigates the selection of MWEs for graded vocabulary lists. Widely used by language teachers and students, these lists recommend a language acquisition sequence to optimize learning efficiency. We automatically generate these lists using difficulty-graded corpora and MWEs extracted based on semantic compositionality. We evaluate these lists on their ability to facilitate text comprehension for learners. Experimental results show that our proposed method generates higher-quality lists than baselines using collocation measures.
Automatic readability assessment (ARA) predicts how difficult it is for the reader to understand a text. While ARA has traditionally been performed at the passage level, there has been increasing interest in ARA at the sentence level, given its applications in downstream tasks such as text simplification and language exercise generation. Recent research has suggested the effectiveness of hybrid approaches for ARA, but they have yet to be applied on the sentence level. We present the first study that compares neural and hybrid models for sentence-level ARA. We conducted experiments on graded sentences from the Wall Street Journal (WSJ) and a dataset derived from the OneStopEnglish corpus. Experimental results show that both neural and hybrid models outperform traditional classifiers trained on linguistic features. Hybrid models obtained the best accuracy on both datasets, surpassing the previous best result reported on the WSJ dataset by almost 13% absolute.
This paper investigates the use of standard and non-standard adverbial markers in modern Chinese literature. In Chinese, adverbials can be derived from many adjectives, adverbs and verbs with the suffix “de”. The suffix has a standard and a non-standard written form, both of which are frequently used. Contrastive research on these two competing forms has mostly been qualitative or limited to small text samples. In this first large-scale quantitative study, we present statistics on 346 adverbial types from an 8-million-character text corpus drawn from Chinese literature in the 20th century. We present a semantic analysis of the verbs modified by adverbs with standard and non-standard markers, and a chronological analysis of marker choice among six prominent modern Chinese authors. We show that the non-standard form is more frequently used when the adverbial modifies an emotion verb. Further, we demonstrate that marker choice is correlated to text genre and register, as well as the writing style of the author.
We present a corpus of simulated counselling sessions consisting of speech- and text-based dialogs in Cantonese. Consisting of 152K Chinese characters, the corpus labels the dialog act of both client and counsellor utterances, segments each dialog into stages, and identifies the forward and backward links in the dialog. We analyze the distribution of client and counsellor communicative intentions in the various stages, and discuss significant patterns of the dialog flow.
The readability assessment task aims to assign a difficulty grade to a text. While neural models have recently demonstrated impressive performance, most do not exploit the ordinal nature of the difficulty grades, and make little effort for model initialization to facilitate fine-tuning. We address these limitations with soft labels for ordinal regression, and with model pre-training through prediction of pairwise relative text difficulty. We incorporate these two components into a model based on hierarchical attention networks, and evaluate its performance on both English and Chinese datasets. Experimental results show that our proposed model outperforms competitive neural models and statistical classifiers on most datasets.
Nominalization re-writes a clause as a noun phrase. It requires the transformation of the head verb of the clause into a deverbal noun, and the verb’s modifiers into nominal modifiers. Past research has focused on the selection of deverbal nouns, but has paid less attention to predicting the word positions and word forms for the nominal modifiers. We propose the use of a textual entailment model for clause nominalization. We obtained the best performance by fine-tuning a textual entailment model on this task, outperforming a number of unsupervised approaches using language model scores from a state-of-the-art neural language model.
To promote efficient learning of Chinese characters, pedagogical materials may present not only a single character, but a set of characters that are related in meaning and in written form. This paper investigates automatic construction of these character sets. The proposed model represents a character as averaged word vectors of common words containing the character. It then identifies sets of characters with high semantic similarity through clustering. Human evaluation shows that this representation outperforms direct use of character embeddings, and that the resulting character sets capture distinct semantic ranges.
Amidst rising mental health needs in society, virtual agents are increasingly deployed in counselling. In order to give pertinent advice, counsellors must first gain an understanding of the issues at hand by eliciting sharing from the counsellee. It is thus important for the counsellor chatbot to encourage the user to open up and talk. One way to sustain the conversation flow is to acknowledge the counsellee’s key points by restating them, or probing them further with questions. This paper applies models from two closely related NLP tasks — summarization and question generation — to restatement and question generation in the counselling context. We conducted experiments on a manually annotated dataset of Cantonese post-reply pairs on topics related to loneliness, academic anxiety and test anxiety. We obtained the best performance in both restatement and question generation by fine-tuning BertSum, a state-of-the-art summarization model, with the in-domain manual dataset augmented with a large-scale, automatically mined open-domain dataset.
A text retrieval system for language learning returns reading materials at the appropriate difficulty level for the user. The system typically maintains a learner model on the user’s vocabulary knowledge, and identifies texts that best fit the model. As the user’s language proficiency increases, model updates are necessary to retrieve texts with the corresponding lexical complexity. We investigate an open learner model that allows user modification of its content, and evaluate its effectiveness with respect to the amount of user update effort. We compare this model with the graded approach, in which the system returns texts at the optimal grade. When the user makes at least half of the expected updates to the open learner model, simulation results show that it outperforms the graded approach in retrieving texts that fit user preference for new-word density.
A nominalization uses a deverbal noun to describe an event associated with its underlying verb. Commonly found in academic and formal texts, nominalizations can be difficult to interpret because of ambiguous semantic relations between the deverbal noun and its arguments. Our goal is to interpret nominalizations by generating clausal paraphrases. We address compound nominalizations with both nominal and adjectival modifiers, as well as prepositional phrases. In evaluations on a number of unsupervised methods, we obtained the strongest performance by using a pre-trained contextualized language model to re-rank paraphrase candidates identified by a textual entailment model.
In many languages, adverbials can be derived from words of various parts-of-speech. In Chinese, the derivation may be marked either with the standard adverbial marker DI, or the non-standard marker DE. Since DE also serves double duty as the attributive marker, accurate identification of adverbials requires disambiguation of its syntactic role. As parsers are trained predominantly on texts using the standard adverbial marker DI, they often fail to recognize adverbials suffixed with the non-standard DE. This paper addresses this problem with an unsupervised, rule-based approach for adverbial identification that utilizes dependency tree patterns. Experiment results show that this approach outperforms a masked language model baseline. We apply this approach to analyze standard and non-standard adverbial marker usage in modern Chinese literature.
Virtual agents are increasingly used for delivering health information in general, and mental health assistance in particular. This paper presents a corpus designed for training a virtual counsellor in Cantonese, a variety of Chinese. The corpus consists of a domain-independent subcorpus that supports small talk for rapport building with users, and a domain-specific subcorpus that provides material for a particular area of counselling. The former consists of ELIZA style responses, chitchat expressions, and a dataset of general dialog, all of which are reusable across counselling domains. The latter consists of example user inputs and appropriate chatbot replies relevant to the specific domain. In a case study, we created a chatbot with a domain-specific subcorpus that addressed 25 issues in test anxiety, with 436 inputs solicited from native speakers of Cantonese and 150 chatbot replies harvested from mental health websites. Preliminary evaluations show that Word Mover’s Distance achieved 56% accuracy in identifying the issue in user input, outperforming a number of baselines.
This paper presents the first investigation on using semantic frames to assess text difficulty. Based on Mandarin VerbNet, a verbal semantic database that adopts a frame-based approach, we examine usage patterns of ten verbs in a corpus of graded Chinese texts. We identify a number of characteristics in texts at advanced grades: more frequent use of non-core frame elements; more frequent omission of some core frame elements; increased preference for noun phrases rather than clauses as verb arguments; and more frequent metaphoric usage. These characteristics can potentially be useful for automatic prediction of text readability.
We present an annotation scheme and a dataset of teacher feedback provided for texts written by non-native speakers of English. The dataset consists of student-written sentences in their original and revised versions with teacher feedback provided for the errors. Feedback appears both in the form of open-ended comments and error category tags. We focus on a specific error type, namely linking adverbial (e.g. however, moreover) errors. The dataset has been annotated for two aspects: (i) revision outcome establishing whether the re-written student sentence was correct and (ii) directness, indicating whether teachers provided explicitly the correction in their feedback. This dataset allows for studies around the characteristics of teacher feedback and how these influence students’ revision outcome. We describe the data preparation process and we present initial statistical investigations regarding the effect of different feedback characteristics on revision outcome. These show that open-ended comments and mitigating expressions appear in a higher proportion of successful revisions than unsuccessful ones, while directness and metalinguistic terms have no effect. Given that the use of this type of data is relatively unexplored in natural language processing (NLP) applications, we also report some observations and challenges when working with feedback data.
We present the first approach to automatically building resources for academic writing. The aim is to build a writing aid system that automatically edits a text so that it better adheres to the academic style of writing. On top of existing academic resources, such as the Corpus of Contemporary American English (COCA) academic Word List, the New Academic Word List, and the Academic Collocation List, we also explore how to dynamically build such resources that would be used to automatically identify informal or non-academic words or phrases. The resources are compiled using different generic approaches that can be extended for different domains and languages. We describe the evaluation of resources with a system implementation. The system consists of an informal word identification (IWI), academic candidate paraphrase generation, and paraphrase ranking components. To generate candidates and rank them in context, we have used the PPDB and WordNet paraphrase resources. We use the Concepts in Context (CoInCO) “All-Words” lexical substitution dataset both for the informal word identification and paraphrase generation experiments. Our informal word identification component achieves an F-1 score of 82%, significantly outperforming a stratified classifier baseline. The main contribution of this work is a domain-independent methodology to build targeted resources for writing aids.
This paper describes a writing assistance system that helps students improve their academic writing. Given an input text, the system suggests lexical substitutions that aim to incorporate more academic vocabulary. The substitution candidates are drawn from an academic word list and ranked by a masked language model. Experimental results show that lexical formality analysis can improve the quality of the suggestions, in comparison to a baseline that relies on the masked language model only.
While bilingual corpora have been instrumental for machine translation, their utility for training translators has been less explored. We investigate the use of bilingual corpora as pedagogical tools for translation in the technical domain. In a user study, novice translators revised Chinese translations of English patents through bilingual concordancing. Results show that concordancing with an in-domain bilingual corpus can yield greater improvement in translation quality of technical terms than a general-domain bilingual corpus.
A lexical simplification (LS) system substitutes difficult words in a text with simpler ones to make it easier for the user to understand. In the typical LS pipeline, the Substitution Ranking step determines the best substitution out of a set of candidates. Most current systems do not consider the user’s vocabulary proficiency, and always aim for the simplest candidate. This approach may overlook less-simple candidates that the user can understand, and that are semantically closer to the original word. We propose a personalized approach for Substitution Ranking to identify the candidate that is the closest synonym and is non-complex for the user. In experiments on learners of English at different proficiency levels, we show that this approach enhances the semantic faithfulness of the output, at the cost of a relatively small increase in the number of complex words.
A lexical simplification (LS) system aims to substitute complex words with simple words in a text, while preserving its meaning and grammaticality. Despite individual users’ differences in vocabulary knowledge, current systems do not consider these variations; rather, they are trained to find one optimal substitution or ranked list of substitutions for all users. We evaluate the performance of a state-of-the-art LS system on individual learners of English at different proficiency levels, and measure the benefits of using complex word identification (CWI) models to personalize the system. Experimental results show that even a simple personalized CWI model, based on graded vocabulary lists, can help the system avoid some unnecessary simplifications and produce more readable output.
This paper describes a personalized text retrieval algorithm that helps language learners select the most suitable reading material in terms of vocabulary complexity. The user first rates their knowledge of a small set of words, chosen by a graph-based active learning model. The system trains a complex word identification model on this set, and then applies the model to find texts that contain the desired proportion of new, challenging, and familiar vocabulary. In an evaluation on learners of Chinese as a foreign language, we show that this algorithm is effective in identifying simpler texts for low-proficiency learners, and more challenging ones for high-proficiency learners.
This paper reports the first study on automatic generation of distractors for fill-in-the-blank items for learning Chinese vocabulary. We investigate the quality of distractors generated by a number of criteria, including part-of-speech, difficulty level, spelling, word co-occurrence and semantic similarity. Evaluations show that a semantic similarity measure, based on the word2vec model, yields distractors that are significantly more plausible than those generated by baseline methods.
Fill-in-the-blank items are a common form of exercise in computer-assisted language learning systems. To automatically generate an effective item, the system must be able to select a high-quality carrier sentence that illustrates the usage of the target word. Previous approaches for carrier sentence selection have considered sentence length, vocabulary difficulty, the position of the target word and the presence of finite verbs. This paper investigates the utility of word co-occurrence statistics and lexical similarity as selection criteria. In an evaluation on generating fill-in-the-blank items for learning Chinese as a foreign language, we show that these two criteria can improve carrier sentence quality.
This opinion paper proposes the use of parallel treebank as learner corpus. We show how an L1-L2 parallel treebank — i.e., parse trees of non-native sentences, aligned to the parse trees of their target hypotheses — can facilitate retrieval of sentences with specific learner errors. We argue for its benefits, in terms of corpus re-use and interoperability, over a conventional learner corpus annotated with error tags. As a proof of concept, we conduct a case study on word-order errors made by learners of Chinese as a foreign language. We report precision and recall in retrieving a range of word-order error categories from L1-L2 tree pairs annotated in the Universal Dependency framework.
This paper applies parsing technology to the task of syntactic simplification of English sentences, focusing on the identification of text spans that can be removed from a complex sentence. We report the most comprehensive evaluation to-date on this task, using a dataset of sentences that exhibit simplification based on coordination, subordination, punctuation/parataxis, adjectival clauses, participial phrases, and appositive phrases. We train a decision tree with features derived from text span length, POS tags and dependency relations, and show that it significantly outperforms a parser-only baseline.
We present the first study that evaluates both speaker and listener identification for direct speech in literary texts. Our approach consists of two steps: identification of speakers and listeners near the quotes, and dialogue chain segmentation. Evaluation results show that this approach outperforms a rule-based approach that is state-of-the-art on a corpus of literary texts.
We explore the application of a Deep Structured Similarity Model (DSSM) to ranking in lexical simplification. Our results show that the DSSM can effectively capture fine-grained features to perform semantic matching when ranking substitution candidates, outperforming the state-of-the-art on two standard datasets used for the task.
We present a web-based interface that automatically assesses reading difficulty of Chinese texts. The system performs word segmentation, part-of-speech tagging and dependency parsing on the input text, and then determines the difficulty levels of the vocabulary items and grammatical constructions in the text. Furthermore, the system highlights the words and phrases that must be simplified or re-written in order to conform to the user-specified target difficulty level. Evaluation results show that the system accurately identifies the vocabulary level of 89.9% of the words, and detects grammar points at 0.79 precision and 0.83 recall.
This article proposes a Universal Dependency Annotation Scheme for Mandarin Chinese, including POS tags and dependency analysis. We identify cases of idiosyncrasy of Mandarin Chinese that are difficult to fit into the current schema which has mainly been based on the descriptions of various Indo-European languages. We discuss differences between our scheme and those of the Stanford Chinese Dependencies and the Chinese Dependency Treebank.
We propose a scheme for annotating direct speech in literary texts, based on the Text Encoding Initiative (TEI) and the coreference annotation guidelines from the Message Understanding Conference (MUC). The scheme encodes the speakers and listeners of utterances in a text, as well as the quotative verbs that reports the utterances. We measure inter-annotator agreement on this annotation task. We then present statistics on a manually annotated corpus that consists of books from the New Testament. Finally, we visualize the corpus as a conversational network.
We present a dependency treebank of the Chinese Buddhist Canon, which contains 1,514 texts with about 50 million Chinese characters. The treebank was created by an automatic parser trained on a smaller treebank, containing four manually annotated sutras (Lee and Kong, 2014). We report results on word segmentation, part-of-speech tagging and dependency parsing, and discuss challenges posed by the processing of medieval Chinese. In a case study, we exploit the treebank to examine verbs frequently associated with Buddha, and to analyze usage patterns of quotative verbs in direct speech. Our results suggest that certain quotative verbs imply status differences between the speaker and the listener.
We present a mobile app that provides a reading environment for learners of Chinese as a foreign language. The app includes a text database that offers over 500K articles from Chinese Wikipedia. These articles have been word-segmented; each word is linked to its entry in a Chinese-English dictionary, and to automatically-generated review exercises. The app estimates the reading proficiency of the user based on a “to-learn” list of vocabulary items. It automatically constructs and maintains this list by tracking the user’s dictionary lookup behavior and performance in review exercises. When a user searches for articles to read, search results are filtered such that the proportion of unknown words does not exceed a user-specified threshold.
We present a browser-based editor for simplifying English text. Given an input sentence, the editor performs both syntactic and lexical simplification. It splits a complex sentence into shorter ones, and suggests word substitutions in drop-down lists. The user can choose the best substitution from the list, undo any inappropriate splitting, and further edit the sentence as necessary. A significant novelty is that the system accepts a customized vocabulary list for a target reader population. It identifies all words in the text that do not belong to the list, and attempts to substitute them with words from the list, thus producing a text tailored for the targeted readers.
We have recently converted a dependency treebank, consisting of ancient Greek and Latin texts, from one annotation scheme to another that was independently designed. This paper makes two observations about this conversion process. First, we show that, despite significant surface differences between the two treebanks, a number of straightforward transformation rules yield a substantial level of compatibility between them, giving evidence for their sound design and high quality of annotation. Second, we analyze some linguistic annotations that require further disambiguation, proposing some simple yet effective machine learning methods.
In this paper, we discuss techniques to combine an interlingua translation framework with phrase-based statistical methods, for translation from Chinese into English. Our goal is to achieve high-quality translation, suitable for use in language tutoring applications. We explore these ideas in the context of a flight domain, for which we have a large corpus of English queries, obtained from users interacting with a dialogue system. Our techniques exploit a pre-existing English-to-Chinese translation system to automatically produce a synthetic bilingual corpus. Several experiments were conducted combining linguistic and statistical methods, and manual evaluation was conducted for a set of 460 Chinese sentences. The best performance achieved an “adequate” or better analysis (3 or above rating) on nearly 94% of the 409 parsable subset. Using a Rover scheme to combine four systems resulted in an “adequate or better” rating for 88% of all the utterances.