A common approach to automatically assigning diagnostic and procedural clinical codes to health records is to solve the task as a multi-label classification problem. Difficulties associated with this task stem from domain knowledge requirements, long document texts, large and imbalanced label space, reflecting the breadth and dependencies between medical diagnoses and procedures. Decisions in the healthcare domain also need to demonstrate sound reasoning, both when they are correct and when they are erroneous. Existing works address some of these challenges by incorporating external knowledge, which can be encoded into a graph-structured format. Incorporating graph structures on the output label space or between the input document and output label spaces have shown promising results in medical codes classification. Limited focus has been put on utilizing graph-based representation on the input document space. To partially bridge this gap, we represent clinical texts as graph-structured data through the UMLS Metathesaurus; we explore implicit graph representation through pre-trained knowledge graph embeddings and explicit domain-knowledge guided encoding of document concepts and relational information through graph neural networks. Our findings highlight the benefits of pre-trained knowledge graph embeddings in understanding model’s attention-based reasoning. In contrast, transparent domain knowledge guidance in graph encoder approaches is overshadowed by performance loss. Our qualitative analysis identifies limitations that contribute to prediction errors.

Pretrained language models (PLMs) form the basis of most state-of-the-art NLP technologies. Nevertheless, they are essentially black boxes: Humans do not have a clear understanding of what knowledge is encoded in different parts of the models, especially in individual neurons. A contrast is in computer vision, where feature visualization provides a decompositional interpretability technique for neurons of vision models. Activation maximization is used to synthesize inherently interpretable visual representations of the information encoded in individual neurons. Our work is inspired by this but presents a cautionary tale on the interpretability of single neurons, based on the first large-scale attempt to adapt activation maximization to NLP, and, more specifically, large PLMs. We propose feature textualization, a technique to produce dense representations of neurons in the PLM word embedding space. We apply feature textualization to the BERT model to investigate whether the knowledge encoded in individual neurons can be interpreted and symbolized. We find that the produced representations can provide insights about the knowledge encoded in individual neurons, but that individual neurons do not represent clear-cut symbolic units of language such as words. Additionally, we use feature textualization to investigate how many neurons are needed to encode words in BERT.

Dense passage retrieval models have become state-of-the-art for information retrieval on many Open-domain Question Answering (ODQA) datasets. However, most of these models rely on supervision obtained from the ODQA datasets, which hinders their performance in a low-resource setting. Recently, retrieval-augmented language models have been proposed to improve both zero-shot and supervised information retrieval. However, these models have pre-training tasks that are agnostic to the target task of passage retrieval. In this work, we propose Retrieval Augmented Auto-encoding of Questions for zero-shot dense information retrieval. Unlike other pre-training methods, our pre-training method is built for target information retrieval, thereby making the pre-training more efficient. Our method consists of a dense IR model for encoding questions and retrieving documents during training and a conditional language model that maximizes the question’s likelihood by marginalizing over retrieved documents. As a by-product, we can use this conditional language model for zero-shot question generation from documents. We show that the IR model obtained through our method improves the current state-of-the-art of zero-shot dense information retrieval, and we improve the results even further by training on a synthetic corpus created by zero-shot question generation.

Relation extraction in the biomedical domain is challenging due to the lack of labeled data and high annotation costs, needing domain experts. Distant supervision is commonly used to tackle the scarcity of annotated data by automatically pairing knowledge graph relationships with raw texts. Such a pipeline is prone to noise and has added challenges to scale for covering a large number of biomedical concepts. We investigated existing broad-coverage distantly supervised biomedical relation extraction benchmarks and found a significant overlap between training and test relationships ranging from 26% to 86%. Furthermore, we noticed several inconsistencies in the data construction process of these benchmarks, and where there is no train-test leakage, the focus is on interactions between narrower entity types. This work presents a more accurate benchmark MedDistant19 for broad-coverage distantly supervised biomedical relation extraction that addresses these shortcomings and is obtained by aligning the MEDLINE abstracts with the widely used SNOMED Clinical Terms knowledge base. Lacking thorough evaluation with domain-specific language models, we also conduct experiments validating general domain relation extraction findings to biomedical relation extraction.

Despite the advances in digital healthcare systems offering curated structured knowledge, much of the critical information still lies in large volumes of unlabeled and unstructured clinical texts. These texts, which often contain protected health information (PHI), are exposed to information extraction tools for downstream applications, risking patient identification. Existing works in de-identification rely on using large-scale annotated corpora in English, which often are not suitable in real-world multilingual settings. Pre-trained language models (LM) have shown great potential for cross-lingual transfer in low-resource settings. In this work, we empirically show the few-shot cross-lingual transfer property of LMs for named entity recognition (NER) and apply it to solve a low-resource and real-world challenge of code-mixed (Spanish-Catalan) clinical notes de-identification in the stroke domain. We annotate a gold evaluation dataset to assess few-shot setting performance where we only use a few hundred labeled examples for training. Our model improves the zero-shot F1-score from 73.7% to 91.2% on the gold evaluation set when adapting Multilingual BERT (mBERT) (CITATION) from the MEDDOCAN (CITATION) corpus with our few-shot cross-lingual target corpus. When generalized to an out-of-sample test set, the best model achieves a human-evaluation F1-score of 97.2%.

Temporal knowledge graph completion (TKGC) has become a popular approach for reasoning over the event and temporal knowledge graphs, targeting the completion of knowledge with accurate but missing information. In this context, tensor decomposition has successfully modeled interactions between entities and relations. Their effectiveness in static knowledge graph completion motivates us to introduce Time-LowFER, a family of parameter-efficient and time-aware extensions of the low-rank tensor factorization model LowFER. Noting several limitations in current approaches to represent time, we propose a cycle-aware time-encoding scheme for time features, which is model-agnostic and offers a more generalized representation of time. We implement our methods in a unified temporal knowledge graph embedding framework, focusing on time-sensitive data processing. The experiments show that our proposed methods perform on par or better than the state-of-the-art semantic matching models on two benchmarks.

There has been a significant progress in the field of Extractive Question Answering (EQA) in the recent years. However, most of them are reliant on annotations of answer-spans in the corresponding passages. In this work, we address the problem of EQA when no annotations are present for the answer span, i.e., when the dataset contains only questions and corresponding passages. Our method is based on auto-encoding of the question that performs a question answering task during encoding and a question generation task during decoding. We show that our method performs well in a zero-shot setting and can provide an additional loss to boost performance for EQA.

Recent advances in deep transformer models have achieved state-of-the-art in several natural language processing (NLP) tasks, whereas named entity recognition (NER) has traditionally benefited from long-short term memory (LSTM) networks. In this work, we present a Transformers based Transfer Learning framework for Named Entity Recognition (T2NER) created in PyTorch for the task of NER with deep transformer models. The framework is built upon the Transformers library as the core modeling engine and supports several transfer learning scenarios from sequential transfer to domain adaptation, multi-task learning, and semi-supervised learning. It aims to bridge the gap between the algorithmic advances in these areas by combining them with the state-of-the-art in transformer models to provide a unified platform that is readily extensible and can be used for both the transfer learning research in NER, and for real-world applications. The framework is available at: https://github.com/suamin/t2ner.

Fact triples are a common form of structured knowledge used within the biomedical domain. As the amount of unstructured scientific texts continues to grow, manual annotation of these texts for the task of relation extraction becomes increasingly expensive. Distant supervision offers a viable approach to combat this by quickly producing large amounts of labeled, but considerably noisy, data. We aim to reduce such noise by extending an entity-enriched relation classification BERT model to the problem of multiple instance learning, and defining a simple data encoding scheme that significantly reduces noise, reaching state-of-the-art performance for distantly-supervised biomedical relation extraction. Our approach further encodes knowledge about the direction of relation triples, allowing for increased focus on relation learning by reducing noise and alleviating the need for joint learning with knowledge graph completion.

Contextualized word embeddings provide better initialization for neural networks that deal with various natural language understanding (NLU) tasks including Question Answering (QA) and more recently, Question Generation(QG). Apart from providing meaningful word representations, pre-trained transformer models (Vaswani et al., 2017), such as BERT (Devlin et al., 2019) also provide self-attentions which encode syntactic information that can be probed for dependency parsing (Hewitt and Manning, 2019) and POStagging (Coenen et al., 2019). In this paper, we show that the information from selfattentions of BERT are useful for language modeling of questions conditioned on paragraph and answer phrases. To control the attention span, we use semi-diagonal mask and utilize a shared model for encoding and decoding, unlike sequence-to-sequence. We further employ copy-mechanism over self-attentions to acheive state-of-the-art results for Question Generation on SQuAD v1.1 (Rajpurkar et al., 2016).

Multilingual, inflectional corpora are a scarce resource in the NLP community, especially corpora with annotated morpheme boundaries. We are evaluating a generated, multilingual inflectional corpus with morpheme boundaries, generated from the English Wiktionary (Metheniti and Neumann, 2018), against the largest, multilingual, high-quality inflectional corpus of the UniMorph project (Kirov et al., 2018). We confirm that the generated Wikinflection corpus is not of such quality as UniMorph, but we were able to extract a significant amount of words from the intersection of the two corpora. Our Wikinflection corpus benefits from the morpheme segmentations of Wiktionary/Wikinflection and from the manually-evaluated morphological feature tags of the UniMorph project, and has 216K lemmas and 5.4M word forms, in a total of 68 languages.

In the following, we describe our system developed for the Semeval2019 Task 8. We fine-tuned a BERT checkpoint on the qatar living forum dump and used this checkpoint to train a number of models. Our hand-in for subtask A consists of a fine-tuned classifier from this BERT checkpoint. For subtask B, we first have a classifier deciding whether a comment is factual or non-factual. If it is factual, we retrieve intra-forum evidence and using this evidence, have a classifier deciding the comment’s veracity. We trained this classifier on ratings which we crawled from qatarliving.com

This paper contains our system description for the second Fact Extraction and VERification (FEVER) challenge. We propose a two-staged sentence selection strategy to account for examples in the dataset where evidence is not only conditioned on the claim, but also on previously retrieved evidence. We use a publicly available document retrieval module and have fine-tuned BERT checkpoints for sentence se- lection and as the entailment classifier. We report a FEVER score of 68.46% on the blind testset.

Code-Mixing (CM) is the phenomenon of alternating between two or more languages which is prevalent in bi- and multi-lingual communities. Most NLP applications today are still designed with the assumption of a single interaction language and are most likely to break given a CM utterance with multiple languages mixed at a morphological, phrase or sentence level. For example, popular commercial search engines do not yet fully understand the intents expressed in CM queries. As a first step towards fostering research which supports CM in NLP applications, we systematically crowd-sourced and curated an evaluation dataset for factoid question answering in three CM languages - Hinglish (Hindi+English), Tenglish (Telugu+English) and Tamlish (Tamil+English) which belong to two language families (Indo-Aryan and Dravidian). We share the details of our data collection process, techniques which were used to avoid inducing lexical bias amongst the crowd workers and other CM specific linguistic properties of the dataset. Our final dataset, which is available freely for research purposes, has 1,694 Hinglish, 2,848 Tamlish and 1,391 Tenglish factoid questions and their answers. We discuss the techniques used by the participants for the first edition of this ongoing challenge.

We present a visualisation tool which aims to illuminate the inner workings of an LSTM model for question answering. It plots heatmaps of neurons’ firings and allows a user to check the dependency between neurons and manual features. The system possesses an interactive web-interface and can be adapted to other models and domains.

We present LightRel, a lightweight and fast relation classifier. Our goal is to develop a high baseline for different relation extraction tasks. By defining only very few data-internal, word-level features and external knowledge sources in the form of word clusters and word embeddings, we train a fast and simple linear classifier

This paper investigates neural character-based morphological tagging for languages with complex morphology and large tag sets. Character-based approaches are attractive as they can handle rarely- and unseen words gracefully. We evaluate on 14 languages and observe consistent gains over a state-of-the-art morphological tagger across all languages except for English and French, where we match the state-of-the-art. We compare two architectures for computing character-based word vectors using recurrent (RNN) and convolutional (CNN) nets. We show that the CNN based approach performs slightly worse and less consistently than the RNN based approach. Small but systematic gains are observed when combining the two architectures by ensembling.

We have developed a new OSGi-based platform for Named Entity Recognition (NER) which uses a voting strategy to combine the results produced by several existing NER systems (currently OpenNLP, LingPipe and Stanford). The different NER systems have been systematically decomposed and modularized into the same pipeline of preprocessing components in order to support a flexible selection and ordering of the NER processing flow. This high modular and component-based design supports the possibility to setup different constellations of chained processing steps including alternative voting strategies for combining the results of parallel running components.

In this paper, we study the problem of finding named entities in the Arabic text. For this task we present the development of our pipeline software for Arabic named entity recognition (ARNE), which includes tokenization, morphological analysis, Buckwalter transliteration, part of speech tagging and named entity recognition of person, location and organisation named entities. In our first attempt to recognize named entites, we have used a simple, fast and language independent gazetteer lookup approach. In our second attempt, we have used the morphological analysis provided by our pipeline to remove affixes and observed hence an improvement in our performance. The pipeline presented in this paper, can be used in future as a basis for a named entity recognition system that recognized named entites not only using gazetteers, but also making use of morphological information and part of speech tagging.

The IDEX system is a prototype of an interactive dynamic Information Extraction (IE) system. A user of the system expresses an information request in the form of a topic description, which is used for an initial search in order to retrieve a relevant set of documents. On basis of this set of documents, unsupervised relation extraction and clustering is done by the system. The results of these operations can then be interactively inspected by the user. In this paper we describe the relation extraction and clustering components of the IDEX system. Preliminary evaluation results of these components are presented and an overview is given of possible enhancements to improve the relation extraction and clustering components.

This paper presents the QALL-ME benchmark, a multilingual resource of annotated spoken requests in the tourism domain, freely available for research purposes. The languages currently involved in the project are Italian, English, Spanish and German. It introduces a semantic annotation scheme for spoken information access requests, specifically derived from Question Answering (QA) research. In addition to pragmatic and semantic annotations, we propose three QA-based annotation levels: the Expected Answer Type, the Expected Answer Quantifier and the Question Topical Target of a request, to fully capture the content of a request and extract the sought-after information. The QALL-ME benchmark is developed under the EU-FP6 QALL-ME project which aims at the realization of a shared and distributed infrastructure for Question Answering (QA) systems on mobile devices (e.g. mobile phones). Questions are formulated by the users in free natural language input, and the system returns the actual sequence of words which constitutes the answer from a collection of information sources (e.g. documents, databases). Within this framework, the benchmark has the twofold purpose of training machine learning based applications for QA, and testing their actual performance with a rapid turnaround in controlled laboratory setting.

We describe a method for the automatic extraction of a Stochastic Lexicalized Tree Insertion Grammar from a linguistically rich HPSG Treebank. The extraction method is strongly guided by HPSG-based head and argument decomposition rules. The tree anchors correspond to lexical labels encoding fine-grained information. The approach has been tested with a German corpus achieving a labeled recall of 77.33% and labeled precision of 78.27%, which is competitive to recent results reported for German parsing using the Negra Treebank.