Gender bias has been extensively studied in both the educational field and the Natural Language Processing (NLP) field, the former using human coding to identify patterns associated with and causes of gender bias in text and the latter to detect, measure and mitigate gender bias in NLP output and models. This work aims to use NLP to facilitate automatic, quantitative analysis of educational text within the framework of a gender bias taxonomy. Analyses of both educational texts and a lexical resource (WordNet) reveal patterns of bias that can inform and aid educators in updating textbooks and lexical resources and in designing assessment items.

Given a sentence and a particular aspect term, aspect-based sentiment analysis (ABSA) aims to predict the sentiment polarity towards this aspect term, which provides fine-grained analysis on sentiment understanding and it has attracted much attention in recent years. In order to achieve a good performance on ABSA, it is important for a model to appropriately encode contextual information, especially identifying salient features and eliminating noise in the context. To make incorrect predictions, most existing approaches employ powerful text encoders to locate important context features, as well as noises that mislead ABSA models. These approaches determine the noise in the text for ABSA by assigning low weights to context features or directly removing them from model input, which runs the risk of computing wrong weights or eliminating important context information. In this paper, we propose to improve ABSA with context denoising, where three types of word-level information are regarded as noise, namely, lexicographic noise, bag-of-words noise, and syntax noise. We utilize diffusion networks to perform the denoising process to gradually eliminate them so as to better predict sentiment polarities for given aspect terms. Our approach uses task-specific noise rather than the standard stochastic Gaussian noise in the diffusion networks. The experimental results on five widely used ABSA datasets demonstrate the validity and effectiveness of our approach.

In many practical scenarios, contents from different modalities are not semantically aligned; for instance, visual and textual information may conflict with each other, resulting in non-compositional expression effects such as irony or humor. Effective modeling and smooth integration of multimodal information are crucial for achieving good understanding of the contrast across modalities. Being focusing on image-text matching, most current studies face challenges in identifying such contrast, leading to limitations in exploring the extended semantics when images and texts do not match. In this paper, we propose an LLM-based approach for learning multimodal contrast following the encoding-decoding paradigm, enhanced by a memory module with reinforced contrast recognition, and use a series of tasks that have the nature of multimodal contrast to verify our approach. The memory module learns the integration between visual and textual features with trainable memory vectors and the reinforced contrast recognition uses self-rejection sampling to optimize the memory to further enhance learning multimodal contrast. The resulted information, accompanied with visual and text features, is finally fed into the LLM to predict corresponding labels. We experiment our approach on four English and Chinese benchmark datasets, where it outperforms strong baselines and state-of-the-art studies.

Recent progress in large language models (LLMs) has marked a notable milestone in the field of artificial intelligence. The conventional evaluation of LLMs primarily relies on existing tasks and benchmarks, raising concerns about test set contamination and the genuine comprehension abilities of LLMs. To address these concerns, we propose to evaluate LLMs by designing new tasks, automatically generating evaluation datasets for the tasks, and conducting detailed error analyses to scrutinize LLMs’ adaptability to new tasks, their sensitivity to prompt variations, and their error tendencies. We investigate the capacity of LLMs to adapt to new but simple tasks, especially when they diverge from the models’ pre-existing knowledge. Our methodology emphasizes the creation of straightforward tasks, facilitating a precise error analysis to uncover the underlying causes of LLM failures. This strategic approach also aims to uncover effective strategies for enhancing LLM performance based on the detailed error analysis of system output.

Remote patient care provides opportunities for expanding medical access, saving healthcare costs, and offering on-demand convenient services. In the MEDIQA-M3G 2024 Shared Task, researchers explored solutions for the specific task of dermatological consumer health visual question answering, where user generated queries and images are used as input and a free-text answer response is generated as output. In this novel challenge, eight teams with a total of 48 submissions were evaluated across three language test sets. In this work, we provide a summary of the dataset, as well as results and approaches. We hope that the insights learned here will inspire future research directions that can lead to technology that deburdens clinical workload and improves care.

Automatic detection and correction of medical errors enables a more rigorous validation of medical documentation as well as clinical notes generated by large language models. Such solutions can ensure the accuracy and medical coherence of clinical texts and enhance patient care and health outcomes. The MEDIQA-CORR 2024 shared task focused on detecting and correcting different types of medical errors in clinical texts. Seventeen teams participated in the shared task and experimented with a broad range of approaches and models. In this paper, we describe the MEDIQA-CORR task, datasets, and the participants’ results and methods.

The development of large language models (LLMs) brings significant changes to the field of natural language processing (NLP), enabling remarkable performance in various high-level tasks, such as machine translation, question-answering, dialogue generation, etc., under end-to-end settings without requiring much training data. Meanwhile, fundamental NLP tasks, particularly syntactic parsing, are also essential for language study as well as evaluating the capability of LLMs for instruction understanding and usage. In this paper, we focus on analyzing and improving the capability of current state-of-the-art LLMs on a classic fundamental task, namely constituency parsing, which is the representative syntactic task in both linguistics and natural language processing. We observe that these LLMs are effective in shallow parsing but struggle with creating correct full parse trees. To improve the performance of LLMs on deep syntactic parsing, we propose a three-step approach that firstly prompts LLMs for chunking, then filters out low-quality chunks, and finally adds the remaining chunks to prompts to instruct LLMs for parsing, with later enhancement by chain-of-thought prompting. Experimental results on English and Chinese benchmark datasets demonstrate the effectiveness of our approach on improving LLMs’ performance on constituency parsing.

Dialogue summarization is an important task that requires to generate highlights for a conversation from different aspects (e.g., content of various speakers). While several studies successfully employ large language models (LLMs) and achieve satisfying results, they are limited by using one model at a time or treat it as a black box, which makes it hard to discriminatively learn essential content in a dialogue from different aspects, therefore may lead to anticipation bias and potential loss of information in the produced summaries. In this paper, we propose an LLM-based approach with role-oriented routing and fusion generation to utilize mixture of experts (MoE) for dialogue summarization. Specifically, the role-oriented routing is an LLM-based module that selects appropriate experts to process different information; fusion generation is another LLM-based module to locate salient information and produce finalized dialogue summaries. The proposed approach offers an alternative solution to employing multiple LLMs for dialogue summarization by leveraging their capabilities of in-context processing and generation in an effective manner. We run experiments on widely used benchmark datasets for this task, where the results demonstrate the superiority of our approach in producing informative and accurate dialogue summarization.

Social determinants of health (SDoH) play a critical role in shaping health outcomes, particularly in pediatric populations where interventions can have long-term implications. SDoH are frequently studied in the Electronic Health Record (EHR), which provides a rich repository for diverse patient data. In this work, we present a novel annotated corpus, the Pediatric Social History Annotation Corpus (PedSHAC), and evaluate the automatic extraction of detailed SDoH representations using fine-tuned and in-context learning methods with Large Language Models (LLMs). PedSHAC comprises annotated social history sections from 1,260 clinical notes obtained from pediatric patients within the University of Washington (UW) hospital system. Employing an event-based annotation scheme, PedSHAC captures ten distinct health determinants to encompass living and economic stability, prior trauma, education access, substance use history, and mental health with an overall annotator agreement of 81.9 F1. Our proposed fine-tuning LLM-based extractors achieve high performance at 78.4 F1 for event arguments. In-context learning approaches with GPT-4 demonstrate promise for reliable SDoH extraction with limited annotated examples, with extraction performance at 82.3 F1 for event triggers.

End-to-end Aspect-based Sentiment Analysis (EASA) is a natural language processing (NLP) task that involves extracting aspect terms and identifying the sentiments for them, which provides a fine-grained level of text analysis and thus requires a deep understanding of the running text. Many previous studies leverage advanced text encoders to extract context information and use syntactic information, e.g., the dependency structure of the input sentence, to improve the model performance. However, such models may reach a bottleneck since the dependency structure is not designed to provide semantic information of the text, which is also important for identifying the sentiment and thus leave room for further improvement. Considering that combinatory categorial grammar (CCG) is a formalism that expresses both syntactic and semantic information of a sentence, it has the potential to be beneficial to EASA. In this paper, we propose a novel approach to improve EASA with CCG supertags, which carry the syntactic and semantic information of the associated words and serve as the most important part of the CCG derivation. Specifically, our approach proposes a CCG supertag decoding process to learn the syntactic and semantic information carried by CCG supertags and use the information to guide the attention over the input words so as to identify important contextual information for EASA. Furthermore, a gate mechanism is used in incorporating the weighted contextual information into the backbone EASA decoding process. We evaluate our approach on three publicly available English datasets for EASA, and show that it outperforms strong baselines and achieves state-of-the-art results on all datasets.

Recently, with the chain of thought (CoT) prompting, large language models (LLMs), e.g., GPT-3, have shown strong reasoning ability in several natural language processing tasks such as arithmetic, commonsense, and logical reasoning. However, LLMs with CoT require multi-step prompting and multi-token prediction, which is highly sensitive to individual mistakes and vulnerable to error accumulation. The above issues make the LLMs need the ability to verify the answers. In fact, after inferring conclusions in some thinking decision tasks, people often check them by re-verifying steps to avoid some mistakes. In this paper, we propose and prove that LLMs also have similar self-verification abilities. We take the conclusion obtained by CoT as one of the conditions for solving the original problem. By performing a backward verification of the answers that LLM deduced for itself, we can obtain interpretable answer validation scores to select the candidate answer with the highest score. Experimental results demonstrate that the proposed method can improve the reasoning performance on various arithmetic, commonsense, and logical reasoning datasets. Our code is publicly available at: https://github.com/WENGSYX/Self-Verification.

Taxonomies, which organize domain concepts into hierarchical structures, are crucial for building knowledge systems and downstream applications. As domain knowledge evolves, taxonomies need to be continuously updated to include new concepts. Previous approaches have mainly focused on adding concepts to the leaf nodes of the existing hierarchical tree, which does not fully utilize the taxonomy’s knowledge and is unable to update the original taxonomy structure (usually involving non-leaf nodes). In this paper, we propose a two-stage method called ATTEMPT for taxonomy completion. Our method inserts new concepts into the correct position by finding a parent node and labeling child nodes. Specifically, by combining local nodes with prompts to generate natural sentences, we take advantage of pre-trained language models for hypernym/hyponymy recognition. Experimental results on two public datasets (including six domains) show that ATTEMPT performs best on both taxonomy completion and extension tasks, surpassing existing methods.

Question Answering (QA) is a Natural Language Processing (NLP) task that can measure language and semantics understanding ability, it requires a system not only to retrieve relevant documents from a large number of articles but also to answer corresponding questions according to documents. However, various language styles and sources of human questions and evidence documents form the different embedding semantic spaces, which may bring some errors to the downstream QA task. To alleviate these problems, we propose a framework for enhancing downstream evidence retrieval by generating evidence, aiming at improving the performance of response generation. Specifically, we take the pre-training language model as a knowledge base, storing documents’ information and knowledge into model parameters. With the Child-Tuning approach being designed, the knowledge storage and evidence generation avoid catastrophic forgetting for response generation. Extensive experiments carried out on the multi-documents dataset show that the proposed method can improve the final performance, which demonstrates the effectiveness of the proposed framework.

This paper introduces the approach of VPAI_Lab team’s experiments on BioNLP 2022 shared task 1 Medical Video Classification (MedVidCL). Given an input video, the MedVidCL task aims to correctly classify it into one of three following categories: Medical Instructional, Medical Non-instructional, and Non-medical. Inspired by its dataset construction process, we divide the classification process into two stages. The first stage is to classify videos into medical videos and non-medical videos. In the second stage, for those samples classified as medical videos, we further classify them into instructional videos and non-instructional videos. In addition, we also propose the cross-modal fusion method to solve the video classification, such as fusing the text features (question and subtitles) from the pre-training language models and visual features from image frames. Specifically, we use textual information to concatenate and query the visual information for obtaining better feature representation. Extensive experiments show that the proposed method significantly outperforms the official baseline method by 15.4% in the F1 score, which shows its effectiveness. Finally, the online results show that our method ranks the Top-1 on the online unseen test set. All the experimental codes are open-sourced at https://github.com/Lireanstar/MedVidCL.

The medical conversational system can relieve doctors’ burden and improve healthcare efficiency, especially during the COVID-19 pandemic. However, the existing medical dialogue systems have the problems of weak scalability, insufficient knowledge, and poor controllability. Thus, we propose a medical conversational question-answering (CQA) system based on the knowledge graph, namely MedConQA, which is designed as a pipeline framework to maintain high flexibility. Our system utilizes automated medical procedures, including medical triage, consultation, image-text drug recommendation, and record. Each module has been open-sourced as a tool, which can be used alone or in combination, with robust scalability. Besides, to conduct knowledge-grounded dialogues with users, we first construct a Chinese Medical Knowledge Graph (CMKG) and collect a large-scale Chinese Medical CQA (CMCQA) dataset, and we design a series of methods for reasoning more intellectually. Finally, we use several state-of-the-art (SOTA) techniques to keep the final generated response more controllable, which is further assured by hospital and professional evaluations. We have open-sourced related code, datasets, web pages, and tools, hoping to advance future research.

Relation extraction (RE) is an important natural language processing task that predicts the relation between two given entities, where a good understanding of the contextual information is essential to achieve an outstanding model performance. Among different types of contextual information, the auto-generated syntactic information (namely, word dependencies) has shown its effectiveness for the task. However, most existing studies require modifications to the existing baseline architectures (e.g., adding new components, such as GCN, on the top of an encoder) to leverage the syntactic information. To offer an alternative solution, we propose to leverage syntactic information to improve RE by training a syntax-induced encoder on auto-parsed data through dependency masking. Specifically, the syntax-induced encoder is trained by recovering the masked dependency connections and types in first, second, and third orders, which significantly differs from existing studies that train language models or word embeddings by predicting the context words along the dependency paths. Experimental results on two English benchmark datasets, namely, ACE2005EN and SemEval 2010 Task 8 datasets, demonstrate the effectiveness of our approach for RE, where our approach outperforms strong baselines and achieve state-of-the-art results on both datasets.

Personal attributes represent structured information about a person, such as their hobbies, pets, family, likes and dislikes. We introduce the tasks of extracting and inferring personal attributes from human-human dialogue, and analyze the linguistic demands of these tasks. To meet these challenges, we introduce a simple and extensible model that combines an autoregressive language model utilizing constrained attribute generation with a discriminative reranker. Our model outperforms strong baselines on extracting personal attributes as well as inferring personal attributes that are not contained verbatim in utterances and instead requires commonsense reasoning and lexical inferences, which occur frequently in everyday conversation. Finally, we demonstrate the benefit of incorporating personal attributes in social chit-chat and task-oriented dialogue settings.

We introduce a Masked Segmental Language Model (MSLM) for joint language modeling and unsupervised segmentation. While near-perfect supervised methods have been developed for segmenting human-like linguistic units in resource-rich languages such as Chinese, many of the world’s languages are both morphologically complex, and have no large dataset of “gold” segmentations for supervised training. Segmental Language Models offer a unique approach by conducting unsupervised segmentation as the byproduct of a neural language modeling objective. However, current SLMs are limited in their scalability due to their recurrent architecture. We propose a new type of SLM for use in both unsupervised and lightly supervised segmentation tasks. The MSLM is built on a span-masking transformer architecture, harnessing a masked bidirectional modeling context and attention, as well as adding the potential for model scalability. In a series of experiments, our model outperforms the segmentation quality of recurrent SLMs on Chinese, and performs similarly to the recurrent model on English.

Chinese word segmentation (CWS) and named entity recognition (NER) are two important tasks in Chinese natural language processing. To achieve good model performance on these tasks, existing neural approaches normally require a large amount of labeled training data, which is often unavailable for specific domains such as the Chinese medical domain due to privacy and legal issues. To address this problem, we have developed a Chinese medical corpus named ChiMST which consists of question-answer pairs collected from an online medical healthcare platform and is annotated with word boundary and medical term information. For word boundary, we mainly follow the word segmentation guidelines for the Penn Chinese Treebank (Xia, 2000); for medical terms, we define 9 categories and 18 sub-categories after consulting medical experts. To provide baselines on this corpus, we train existing state-of-the-art models on it and achieve good performance. We believe that the corpus and the baseline systems will be a valuable resource for CWS and NER research on the medical domain.

Aspect-based sentiment analysis (ABSA) aims to predict the sentiment polarity towards a given aspect term in a sentence on the fine-grained level, which usually requires a good understanding of contextual information, especially appropriately distinguishing of a given aspect and its contexts, to achieve good performance. However, most existing ABSA models pay limited attention to the modeling of the given aspect terms and thus result in inferior results when a sentence contains multiple aspect terms with contradictory sentiment polarities. In this paper, we propose to improve ABSA by complementary learning of aspect terms, which serves as a supportive auxiliary task to enhance ABSA by explicitly recovering the aspect terms from each input sentence so as to better understand aspects and their contexts. Particularly, a discriminator is also introduced to further improve the learning process by appropriately balancing the impact of aspect recovery to sentiment prediction. Experimental results on five widely used English benchmark datasets for ABSA demonstrate the effectiveness of our approach, where state-of-the-art performance is observed on all datasets.

As an important task to analyze the semantic structure of a sentence, semantic role labeling (SRL) aims to locate the semantic role (e.g., agent) of noun phrases with respect to a given predicate and thus plays an important role in downstream tasks such as dialogue systems. To achieve a better performance in SRL, a model is always required to have a good understanding of the context information. Although one can use advanced text encoder (e.g., BERT) to capture the context information, extra resources are also required to further improve the model performance. Considering that there are correlations between the syntactic structure and the semantic structure of the sentence, many previous studies leverage auto-generated syntactic knowledge, especially the dependencies, to enhance the modeling of context information through graph-based architectures, where limited attention is paid to other types of auto-generated knowledge. In this paper, we propose map memories to enhance SRL by encoding different types of auto-generated syntactic knowledge (i.e., POS tags, syntactic constituencies, and word dependencies) obtained from off-the-shelf toolkits. Experimental results on two English benchmark datasets for span-style SRL (i.e., CoNLL-2005 and CoNLL-2012) demonstrate the effectiveness of our approach, which outperforms strong baselines and achieves state-of-the-art results on CoNLL-2005.

This paper introduces the approach of Team LingJing’s experiments on SemEval-2022 Task 1 Comparing Dictionaries and Word Embeddings (CODWOE). This task aims at comparing two types of semantic descriptions and including two sub-tasks: the definition modeling and reverse dictionary track. Our team focuses on the reverse dictionary track and adopts the multi-task self-supervised pre-training for multilingual reverse dictionaries. Specifically, the randomly initialized mDeBERTa-base model is used to perform multi-task pre-training on the multilingual training datasets. The pre-training step is divided into two stages, namely the MLM pre-training stage and the contrastive pre-training stage. The experimental results show that the proposed method has achieved good performance in the reverse dictionary track, where we rank the 1-st in the Sgns targets of the EN and RU languages. All the experimental codes are open-sourced at https://github.com/WENGSYX/Semeval.

This paper presents the results and main findings of our system on SemEval-2022 Task 3 Presupposed Taxonomies: Evaluating Neural Network Semantics (PreTENS). This task aims at semantic competence with specific attention on the evaluation of language models, which is a task with respect to the recognition of appropriate taxonomic relations between two nominal arguments. Two sub-tasks including binary classification and regression are designed for the evaluation. For the classification sub-task, we adopt the DeBERTa-v3 pre-trained model for fine-tuning datasets of different languages. Due to the small size of the training datasets of the regression sub-task, we transfer the knowledge of classification model (i.e., model parameters) to the regression task. The experimental results show that the proposed method achieves the best results on both sub-tasks. Meanwhile, we also report negative results of multiple training strategies for further discussion. All the experimental codes are open-sourced at https://github.com/WENGSYX/Semeval.

Dependency parsing is an important fundamental natural language processing task which analyzes the syntactic structure of an input sentence by illustrating the syntactic relations between words. To improve dependency parsing, leveraging existing dependency parsers and extra data (e.g., through semi-supervised learning) has been demonstrated to be effective, even though the final parsers are trained on inaccurate (but massive) data. In this paper, we propose a frustratingly easy approach to improve graph-based dependency parsing, where a structure-aware encoder is pre-trained on auto-parsed data by predicting the word dependencies and then fine-tuned on gold dependency trees, which differs from the usual pre-training process that aims to predict the context words along dependency paths. Experimental results and analyses demonstrate the effectiveness and robustness of our approach to benefit from the data (even with noise) processed by different parsers, where our approach outperforms strong baselines under different settings with different dependency standards and model architectures used in pre-training and fine-tuning. More importantly, further analyses find that only 2K auto-parsed sentences are required to obtain improvement when pre-training vanilla BERT-large based parser without requiring extra parameters.

Summarization is an important natural language processing (NLP) task in identifying key information from text. For conversations, the summarization systems need to extract salient contents from spontaneous utterances by multiple speakers. In a special task-oriented scenario, namely medical conversations between patients and doctors, the symptoms, diagnoses, and treatments could be highly important because the nature of such conversation is to find a medical solution to the problem proposed by the patients. Especially consider that current online medical platforms provide millions of public available conversations between real patients and doctors, where the patients propose their medical problems and the registered doctors offer diagnosis and treatment, a conversation in most cases could be too long and the key information is hard to be located. Therefore, summarizations to the patients’ problems and the doctors’ treatments in the conversations can be highly useful, in terms of helping other patients with similar problems have a precise reference for potential medical solutions. In this paper, we focus on medical conversation summarization, using a dataset of medical conversations and corresponding summaries which were crawled from a well-known online healthcare service provider in China. We propose a hierarchical encoder-tagger model (HET) to generate summaries by identifying important utterances (with respect to problem proposing and solving) in the conversations. For the particular dataset used in this study, we show that high-quality summaries can be generated by extracting two types of utterances, namely, problem statements and treatment recommendations. Experimental results demonstrate that HET outperforms strong baselines and models from previous studies, and adding conversation-related features can further improve system performance.

Chinese word segmentation (CWS) and part-of-speech (POS) tagging are two fundamental tasks for Chinese language processing. Previous studies have demonstrated that jointly performing them can be an effective one-step solution to both tasks and this joint task can benefit from a good modeling of contextual features such as n-grams. However, their work on modeling such contextual features is limited to concatenating the features or their embeddings directly with the input embeddings without distinguishing whether the contextual features are important for the joint task in the specific context. Therefore, their models for the joint task could be misled by unimportant contextual information. In this paper, we propose a character-based neural model for the joint task enhanced by multi-channel attention of n-grams. In the attention module, n-gram features are categorized into different groups according to several criteria, and n-grams in each group are weighted and distinguished according to their importance for the joint task in the specific context. To categorize n-grams, we try two criteria in this study, i.e., n-gram frequency and length, so that n-grams having different capabilities of carrying contextual information are discriminatively learned by our proposed attention module. Experimental results on five benchmark datasets for CWS and POS tagging demonstrate that our approach outperforms strong baseline models and achieves state-of-the-art performance on all five datasets.

Statistical significance testing centered on p-values is commonly used to compare NLP system performance, but p-values alone are insufficient because statistical significance differs from practical significance. The latter can be measured by estimating effect size. In this pa-per, we propose a three-stage procedure for comparing NLP system performance and provide a toolkit, NLPStatTest, that automates the process. Users can upload NLP system evaluation scores and the toolkit will analyze these scores, run appropriate significance tests, estimate effect size, and conduct power analysis to estimate Type II error. The toolkit provides a convenient and systematic way to compare NLP system performance that goes beyond statistical significance testing.

Contextual features always play an important role in Chinese word segmentation (CWS). Wordhood information, being one of the contextual features, is proved to be useful in many conventional character-based segmenters. However, this feature receives less attention in recent neural models and it is also challenging to design a framework that can properly integrate wordhood information from different wordhood measures to existing neural frameworks. In this paper, we therefore propose a neural framework, WMSeg, which uses memory networks to incorporate wordhood information with several popular encoder-decoder combinations for CWS. Experimental results on five benchmark datasets indicate the memory mechanism successfully models wordhood information for neural segmenters and helps WMSeg achieve state-of-the-art performance on all those datasets. Further experiments and analyses also demonstrate the robustness of our proposed framework with respect to different wordhood measures and the efficiency of wordhood information in cross-domain experiments.

Chinese word segmentation (CWS) and part-of-speech (POS) tagging are important fundamental tasks for Chinese language processing, where joint learning of them is an effective one-step solution for both tasks. Previous studies for joint CWS and POS tagging mainly follow the character-based tagging paradigm with introducing contextual information such as n-gram features or sentential representations from recurrent neural models. However, for many cases, the joint tagging needs not only modeling from context features but also knowledge attached to them (e.g., syntactic relations among words); limited efforts have been made by existing research to meet such needs. In this paper, we propose a neural model named TwASP for joint CWS and POS tagging following the character-based sequence labeling paradigm, where a two-way attention mechanism is used to incorporate both context feature and their corresponding syntactic knowledge for each input character. Particularly, we use existing language processing toolkits to obtain the auto-analyzed syntactic knowledge for the context, and the proposed attention module can learn and benefit from them although their quality may not be perfect. Our experiments illustrate the effectiveness of the two-way attentions for joint CWS and POS tagging, where state-of-the-art performance is achieved on five benchmark datasets.

Medical conversation is a central part of medical care. Yet, the current state and quality of medical conversation is far from perfect. Therefore, a substantial amount of research has been done to obtain a better understanding of medical conversation and to address its practical challenges and dilemmas. In line with this stream of research, we have developed a multi-layer structure annotation scheme to analyze medical conversation, and are using the scheme to construct a corpus of naturally occurring medical conversation in Chinese pediatric primary care setting. Some of the preliminary findings are reported regarding 1) how a medical conversation starts, 2) where communication problems tend to occur, and 3) how physicians close a conversation. Challenges and opportunities for research on medical conversation with NLP techniques will be discussed.

Constituency parsing is a fundamental and important task for natural language understanding, where a good representation of contextual information can help this task. N-grams, which is a conventional type of feature for contextual information, have been demonstrated to be useful in many tasks, and thus could also be beneficial for constituency parsing if they are appropriately modeled. In this paper, we propose span attention for neural chart-based constituency parsing to leverage n-gram information. Considering that current chart-based parsers with Transformer-based encoder represent spans by subtraction of the hidden states at the span boundaries, which may cause information loss especially for long spans, we incorporate n-grams into span representations by weighting them according to their contributions to the parsing process. Moreover, we propose categorical span attention to further enhance the model by weighting n-grams within different length categories, and thus benefit long-sentence parsing. Experimental results on three widely used benchmark datasets demonstrate the effectiveness of our approach in parsing Arabic, Chinese, and English, where state-of-the-art performance is obtained by our approach on all of them.

Supertagging is conventionally regarded as an important task for combinatory categorial grammar (CCG) parsing, where effective modeling of contextual information is highly important to this task. However, existing studies have made limited efforts to leverage contextual features except for applying powerful encoders (e.g., bi-LSTM). In this paper, we propose attentive graph convolutional networks to enhance neural CCG supertagging through a novel solution of leveraging contextual information. Specifically, we build the graph from chunks (n-grams) extracted from a lexicon and apply attention over the graph, so that different word pairs from the contexts within and across chunks are weighted in the model and facilitate the supertagging accordingly. The experiments performed on the CCGbank demonstrate that our approach outperforms all previous studies in terms of both supertagging and parsing. Further analyses illustrate the effectiveness of each component in our approach to discriminatively learn from word pairs to enhance CCG supertagging.

Question answering (QA) is a challenging task in natural language processing (NLP), especially when it is applied to specific domains. While models trained in the general domain can be adapted to a new target domain, their performance often degrades significantly due to domain mismatch. Alternatively, one can require a large amount of domain-specific QA data, but such data are rare, especially for the medical domain. In this study, we first collect a large-scale Chinese medical QA corpus called ChiMed; second we annotate a small fraction of the corpus to check the quality of the answers; third, we extract two datasets from the corpus and use them for the relevancy prediction task and the adoption prediction task. Several benchmark models are applied to the datasets, producing good results for both tasks.

Natural language inference (NLI) is challenging, especially when it is applied to technical domains such as biomedical settings. In this paper, we propose a hybrid approach to biomedical NLI where different types of information are exploited for this task. Our base model includes a pre-trained text encoder as the core component, and a syntax encoder and a feature encoder to capture syntactic and domain-specific information. Then we combine the output of different base models to form more powerful ensemble models. Finally, we design two conflict resolution strategies when the test data contain multiple (premise, hypothesis) pairs with the same premise. We train our models on the MedNLI dataset, yielding the best performance on the test set of the MEDIQA 2019 Task 1.

This paper describes the COSTA scheme for coding structures and actions in conversation. Informed by Conversation Analysis, the scheme introduces an innovative method for marking multi-layer structural organization of conversation and a structure-informed taxonomy of actions. In addition, we create a corpus of naturally occurring medical conversations, containing 318 video-recorded and manually transcribed pediatric consultations. Based on the annotated corpus, we investigate 1) treatment decision-making process in medical conversations, and 2) effects of physician-caregiver communication behaviors on antibiotic over-prescribing. Although the COSTA annotation scheme is developed based on data from the task-specific domain of pediatric consultations, it can be easily extended to apply to more general domains and other languages.

Conventional word embeddings are trained with specific criteria (e.g., based on language modeling or co-occurrence) inside a single information source, disregarding the opportunity for further calibration using external knowledge. This paper presents a unified framework that leverages pre-learned or external priors, in the form of a regularizer, for enhancing conventional language model-based embedding learning. We consider two types of regularizers. The first type is derived from topic distribution by running LDA on unlabeled data. The second type is based on dictionaries that are created with human annotation efforts. To effectively learn with the regularizers, we propose a novel data structure, trajectory softmax, in this paper. The resulting embeddings are evaluated by word similarity and sentiment classification. Experimental results show that our learning framework with regularization from prior knowledge improves embedding quality across multiple datasets, compared to a diverse collection of baseline methods.

Crowdsourcing has proven to be an effective method for generating labeled data for a range of NLP tasks. However, multiple recent attempts of using crowdsourcing to generate gold-labeled training data for semantic role labeling (SRL) reported only modest results, indicating that SRL is perhaps too difficult a task to be effectively crowdsourced. In this paper, we postulate that while producing SRL annotation does require expert involvement in general, a large subset of SRL labeling tasks is in fact appropriate for the crowd. We present a novel workflow in which we employ a classifier to identify difficult annotation tasks and route each task either to experts or crowd workers according to their difficulties. Our experimental evaluation shows that the proposed approach reduces the workload for experts by over two-thirds, and thus significantly reduces the cost of producing SRL annotation at little loss in quality.

Early detection and treatment of diseases that onset after a patient is admitted to a hospital, such as pneumonia, is critical to improving and reducing costs in healthcare. Previous studies (Tepper et al., 2013) showed that change-of-state events in clinical notes could be important cues for phenotype detection. In this paper, we extend the annotation schema proposed in (Klassen et al., 2014) to mark change-of-state events, diagnosis events, coordination, and negation. After we have completed the annotation, we build NLP systems to automatically identify named entities and medical events, which yield an f-score of 94.7% and 91.8%, respectively.

In this paper, we describe the expansion of the ODIN resource, a database containing many thousands of instances of Interlinear Glossed Text (IGT) for over a thousand languages harvested from scholarly linguistic papers posted to the Web. A database containing a large number of instances of IGT, which are effectively richly annotated and heuristically aligned bitexts, provides a unique resource for bootstrapping NLP tools for resource-poor languages. To make the data in ODIN more readily consumable by tool developers and NLP researchers, we propose a new XML format for IGT, called Xigt. We call the updated release ODIN-II.

Languages change over time and ancient languages have been studied in linguistics and other related fields. A main challenge in this research area is the lack of empirical data; for instance, ancient spoken languages often leave little trace of their linguistic properties. From the perspective of natural language processing (NLP), while the NLP community has created dozens of annotated corpora, very few of them are on ancient languages. As an effort toward bridging the gap, we have created a word segmented and POS tagged corpus for Archaic Chinese using articles from Huainanzi, a book written during China’s Western Han Dynasty (206 BC-9 AD). We then compare this corpus with the Chinese Penn Treebank (CTB), a well-known corpus for Modern Chinese, and report several interesting differences and similarities between the two corpora. Finally, we demonstrate that the CTB can be used to improve the performance of word segmenters and POS taggers for Archaic Chinese, but only through features that have similar behaviors in the two corpora.

Early detection and treatment of diseases that onset after a patient is admitted to a hospital, such as pneumonia, is critical to improving and reducing costs in healthcare. NLP systems that analyze the narrative data embedded in clinical artifacts such as x-ray reports can help support early detection. In this paper, we consider the importance of identifying the change of state for events - in particular, clinical events that measure and compare the multiple states of a patient’s health across time. We propose a schema for event annotation comprised of five fields and create preliminary annotation guidelines for annotators to apply the schema. We then train annotators, measure their performance, and finalize our guidelines. With the complete guidelines, we then annotate a corpus of snippets extracted from chest x-ray reports in order to integrate the annotations as a new source of features for classification tasks.

Syntactic parses can provide valuable information for many NLP tasks, such as machine translation, semantic analysis, etc. However, most of the world's languages do not have large amounts of syntactically annotated corpora available for building parsers. Syntactic projection techniques attempt to address this issue by using parallel corpora between resource-poor and resource-rich languages, bootstrapping the resource-poor language with the syntactic analysis of the resource-rich language. In this paper, we investigate the possibility of using small, parallel, annotated corpora to automatically detect divergent structural patterns between two languages. These patterns can then be used to improve structural projection algorithms, allowing for better performing NLP tools for resource-poor languages, in particular those that may not have large amounts of annotated data necessary for traditional, fully-supervised methods. While this detection process is not exhaustive, we demonstrate that important instances of divergence are picked up with minimal prior knowledge of a given language pair.

Domain adaptation is an important topic for natural language processing. There has been extensive research on the topic and various methods have been explored, including training data selection, model combination, semi-supervised learning. In this study, we propose to use a goodness measure, namely, description length gain (DLG), for domain adaptation for Chinese word segmentation. We demonstrate that DLG can help domain adaptation in two ways: as additional features for supervised segmenters to improve system performance, and also as a similarity measure for selecting training data to better match a test set. We evaluated our systems on the Chinese Penn Treebank version 7.0, which has 1.2 million words from five different genres, and the Chinese Word Segmentation Bakeoff-3 data.

Automatically segmenting and classifying clinical free text into sections is an important first step to automatic information retrieval, information extraction and data mining tasks, as it helps to ground the significance of the text within. In this work we describe our approach to automatic section segmentation of clinical records such as hospital discharge summaries and radiology reports, along with section classification into pre-defined section categories. We apply machine learning to the problems of section segmentation and section classification, comparing a joint (one-step) and a pipeline (two-step) approach. We demonstrate that our systems perform well when tested on three data sets, two for hospital discharge summaries and one for radiology reports. We then show the usefulness of section information by incorporating it in the task of extracting comorbidities from discharge summaries.

Domain adaptation is an important task in order for NLP systems to work well in real applications. There has been extensive research on this topic. In this paper, we address two issues that are related to domain adaptation. The first question is how much genre variation will affect NLP systems' performance. We investigate the effect of genre variation on the performance of three NLP tools, namely, word segmenter, POS tagger, and parser. We choose the Chinese Penn Treebank (CTB) as our corpus. The second question is how one can estimate NLP systems' performance when gold standard on the test data does not exist. To answer the question, we extend the prediction model in (Ravi et al., 2008) to provide prediction for word segmentation and POS tagging as well. Our experiments show that the predicted scores are close to the real scores when tested on the CTB data.

We are in the process of creating a multi-representational and multi-layered treebank for Hindi/Urdu (Palmer et al., 2009), which has three main layers: dependency structure, predicate-argument structure (PropBank), and phrase structure. This paper discusses an important issue in treebank design which is often neglected: the use of empty categories (ECs). All three levels of representation make use of ECs. We make a high-level distinction between two types of ECs, trace and silent, on the basis of whether they are postulated to mark displacement or not. Each type is further refined into several subtypes based on the underlying linguistic phenomena which the ECs are introduced to handle. This paper discusses the stages at which we add ECs to the Hindi/Urdu treebank and why. We investigate methodically the different types of ECs and their role in our syntactic and semantic representations. We also examine our decisions whether or not to coindex each type of ECs with other elements in the representation.

As the data for more and more languages is finding its way into digital form, with an increasing amount of this data being posted to the Web, it has become possible to collect language data from the Web and create large multilingual resources, covering hundreds or even thousands of languages. ODIN, the Online Database of INterlinear text (Lewis, 2006), is such a resource. It currently consists of nearly 200,000 data points for over 1,000 languages, the data for which was harvested from linguistic documents on the Web. We identify a number of issues with language identification for such broad-coverage resources including the lack of training data, ambiguous language names, incomplete language code sets, and incorrect uses of language names and codes. After providing a short overview of existing language code sets maintained by the linguistic community, we discuss what linguists and the linguistic community can do to make the process of language identification easier.