Lexical inference in context (LIiC) is the task of recognizing textual entailment between two very similar sentences, i.e., sentences that only differ in one expression. It can therefore be seen as a variant of the natural language inference task that is focused on lexical semantics. We formulate and evaluate the first approaches based on pretrained language models (LMs) for this task: (i) a few-shot NLI classifier, (ii) a relation induction approach based on handcrafted patterns expressing the semantics of lexical inference, and (iii) a variant of (ii) with patterns that were automatically extracted from a corpus. All our approaches outperform the previous state of the art, showing the potential of pretrained LMs for LIiC. In an extensive analysis, we investigate factors of success and failure of our three approaches.
Center for Information and Language Processing (CIS)（信息与语言处理中心）；LMU Munich（慕尼黑大学）
Combining a pretrained language model (PLM) with textual patterns has been shown to help in both zero- and few-shot settings. For zero-shot performance, it makes sense to design patterns that closely resemble the text seen during self-supervised pretraining because the model has never seen anything else. Supervised training allows for more flexibility. If we allow for tokens outside the PLM's vocabulary, patterns can be adapted more flexibly to a PLM's idiosyncrasies. Contrasting patterns where a "token" can be any continuous vector vs. those where a discrete choice between vocabulary elements…
The University of Melbourne（墨尔本大学）；RIKEN Center for Advanced Intelligence Project (AIP)（理化学研究所高级智能项目中心）；MBZUAI（穆罕默德·本·扎耶德人工智能大学）
We propose a new unsupervised method for lexical substitution using pre-trained language models. Compared to previous approaches that use the generative capability of language models to predict substitutes, our method retrieves substitutes based on the similarity of contextualised and decontextualised word embeddings, i.e. the average contextual representation of a word in multiple contexts. We conduct experiments in English and Italian, and show that our method substantially outperforms strong baselines and establishes a new state-of-the-art without any explicit supervision or fine-tuning. We…
University of Cambridge（剑桥大学）；University of Mannheim（曼海姆大学）
The success of large pretrained language models (LMs) such as BERT and RoBERTa has sparked interest in probing their representations, in order to unveil what types of knowledge they implicitly capture. While prior research focused on morphosyntactic, semantic, and world knowledge, it remains unclear to which extent LMs also derive lexical type-level knowledge from words in context. In this work, we present a systematic empirical analysis across six typologically diverse languages and five different lexical tasks, addressing the following questions: 1) How do different lexical knowledge extract…
Comparing human and LLM proofreading in L2 writing: Impact on lexical and syntactic features
LCR-ADS Lab, Linguistics, University of Oregon（俄勒冈大学语言学系LCR-ADS实验室）；West University of Timisoara（蒂米什瓦拉西大学）；Gyeongin National University of Education（京仁国立教育大学）
Contextualized word embeddings have demonstrated state-of-the-art performance in various natural language processing tasks including those that concern historical semantic change. However, language models such as BERT was trained primarily on contemporary corpus data. To investigate whether training on historical corpus data improves diachronic semantic analysis, we present a pre-trained BERT-based language model, HistBERT, trained on the balanced Corpus of Historical American English. We examine the effectiveness of our approach by comparing the performance of the original BERT and that of Hi…
Center for Information and Language Processing (CIS)（信息与语言处理中心（CIS））；LMU Munich（慕尼黑大学）
We present SherLIiC, a testbed for lexical inference in context (LIiC), consisting of 3985 manually annotated inference rule candidates (InfCands), accompanied by (i) ~960k unlabeled InfCands, and (ii) ~190k typed textual relations between Freebase entities extracted from the large entity-linked corpus ClueWeb09. Each InfCand consists of one of these relations, expressed as a lemmatized dependency path, and two argument placeholders, each linked to one or more Freebase types. Due to our candidate selection process based on strong distributional evidence, SherLIiC is much harder than existing t…
The University of Tokyo（东京大学）；National Institute of Informatics（国立信息学研究所）
Question answering (QA) models for reading comprehension have been demonstrated to exploit unintended dataset biases such as question-context lexical overlap. This hinders QA models from generalizing to under-represented samples such as questions with low lexical overlap. Question generation (QG), a method for augmenting QA datasets, can be a solution for such performance degradation if QG can properly debias QA datasets. However, we discover that recent neural QG models are biased towards generating questions with high lexical overlap, which can amplify the dataset bias. Moreover, our analysi…
TU Darmstadt（达姆施塔特工业大学）；Bar-Ilan University（巴伊兰大学）
We create a new NLI test set that shows the deficiency of state-of-the-art models in inferences that require lexical and world knowledge. The new examples are simpler than the SNLI test set, containing sentences that differ by at most one word from sentences in the training set. Yet, the performance on the new test set is substantially worse across systems trained on SNLI, demonstrating that these systems are limited in their generalization ability, failing to capture many simple inferences.
We investigate neg(ation)-raising inferences, wherein negation on a predicate can be interpreted as though in that predicate's subordinate clause. To do this, we collect a large-scale dataset of neg-raising judgments for effectively all English clause-embedding verbs and develop a model to jointly induce the semantic types of verbs and their subordinate clauses and the relationship of these types to neg-raising inferences. We find that some neg-raising inferences are attributable to properties of particular predicates, while others are attributable to subordinate clause structure.
This paper presents a partial solution to a component of the problem of lexical choice: choosing the synonym most typical, or expected, in context. We apply a new statistical approach to representing the context of a word through lexical co-occurrence networks. The implementation was trained and evaluated on a large corpus, and results show that the inclusion of second-order co-occurrence relations improves the performance of our implemented lexical choice program.
Stanford University（斯坦福大学）；Allen Institute for AI（艾伦人工智能研究所）
We address whether neural models for Natural Language Inference (NLI) can learn the compositional interactions between lexical entailment and negation, using four methods: the behavioral evaluation methods of (1) challenge test sets and (2) systematic generalization tasks, and the structural evaluation methods of (3) probes and (4) interventions. To facilitate this holistic evaluation, we present Monotonicity NLI (MoNLI), a new naturalistic dataset focused on lexical entailment and negation. In our behavioral evaluations, we find that models trained on general-purpose NLI datasets fail systema…
Natural language understanding applications such as interactive planning and face-to-face translation require extensive inferencing. Many of these inferences are based on the meaning of particular open class words. Providing a representation that can support such lexically-based inferences is a primary concern of lexical semantics. The representation language of first order logic has well-understood semantics and a multitude of inferencing systems have been implemented for it. Thus it is a prime candidate to serve as a lexical semantics representation. However, we argue that FOL, although a go…
Insight SFI Research Centre for Data Analytics（Insight SFI数据分析研究中心）；School of ICS（信息与计算科学学院）；University College Dublin（都柏林大学学院）
This paper presents a new method for automatically detecting words with lexical gender in large-scale language datasets. Currently, the evaluation of gender bias in natural language processing relies on manually compiled lexicons of gendered expressions, such as pronouns ('he', 'she', etc.) and nouns with lexical gender ('mother', 'boyfriend', 'policewoman', etc.). However, manual compilation of such lists can lead to static information if they are not periodically updated and often involve value judgments by individual annotators and researchers. Moreover, terms not included in the list fall …
In neural network models of language, words are commonly represented using context-invariant representations (word embeddings) which are then put in context in the hidden layers. Since words are often ambiguous, representing the contextually relevant information is not trivial. We investigate how an LSTM language model deals with lexical ambiguity in English, designing a method to probe its hidden representations for lexical and contextual information about words. We find that both types of information are represented to a large extent, but also that there is room for improvement for contextua…
Interpretation of natural language is inherently context-sensitive. Most words in natural language are ambiguous and their meanings are heavily dependent on the linguistic context in which they are used. The study of lexical semantics can not be separated from the notion of context. This paper takes a contextual approach to lexical semantics and studies the linguistic context of lexical atoms, or "sticky" phrases such as "hot dog". Since such lexical atoms may occur frequently in unrestricted natural language text, recognizing them is crucial for understanding naturally-occurring text. The pap…
Islamic Science University of Malaysia（马来西亚伊斯兰科学大学）；Zarqa University（扎尔卡大学）
Question processing is a fundamental step in a question answering (QA) application, and its quality impacts the performance of QA application. The major challenging issue in processing question is how to extract semantic of natural language questions (NLQs). A human language is ambiguous. Ambiguity may occur at two levels; lexical and syntactic. In this paper, we propose a new approach for resolving lexical ambiguity problem by integrating context knowledge and concepts knowledge of a domain, into shallow natural language processing (SNLP) techniques. Concepts knowledge is modeled using ontolo…
Multiple (simple) context-free tree grammars are investigated, where "simple" means "linear and nondeleting". Every multiple context-free tree grammar that is finitely ambiguous can be lexicalized; i.e., it can be transformed into an equivalent one (generating the same tree language) in which each rule of the grammar contains a lexical symbol. Due to this transformation, the rank of the nonterminals increases at most by 1, and the multiplicity (or fan-out) of the grammar increases at most by the maximal rank of the lexical symbols; in particular, the multiplicity does not increase when all lex…
A word having multiple senses in a text introduces the lexical semantic task to find out which particular sense is appropriate for the given context. One such task is Word sense disambiguation which refers to the identification of the most appropriate meaning of the polysemous word in a given context using computational algorithms. The language processing research in Hindi, the official language of India, and other Indian languages is restricted by unavailability of the standard corpus. For Hindi word sense disambiguation also, the large corpus is not available. In this work, we prepared the t…
Contextualized embeddings such as BERT can serve as strong input representations to NLP tasks, outperforming their static embeddings counterparts such as skip-gram, CBOW and GloVe. However, such embeddings are dynamic, calculated according to a sentence-level context, which limits their use in lexical semantics tasks. We address this issue by making use of dynamic embeddings as word representations in training static embeddings, thereby leveraging their strong representation power for disambiguating context information. Results show that this method leads to improvements over traditional stati…
Automatic phylogenetic inference plays an increasingly important role in computational historical linguistics. Most pertinent work is currently based on expert cognate judgments. This limits the scope of this approach to a small number of well-studied language families. We used machine learning techniques to compile data suitable for phylogenetic inference from the ASJP database, a collection of almost 7,000 phonetically transcribed word lists over 40 concepts, covering two third of the extant world-wide linguistic diversity. First, we estimated Pointwise Mutual Information scores between soun…
Purdue University（普渡大学）；University of Chicago（芝加哥大学）
Models trained to estimate word probabilities in context have become ubiquitous in natural language processing. How do these models use lexical cues in context to inform their word probabilities? To answer this question, we present a case study analyzing the pre-trained BERT model with tests informed by semantic priming. Using English lexical stimuli that show priming in humans, we find that BERT too shows "priming," predicting a word with greater probability when the context includes a related word versus an unrelated one. This effect decreases as the amount of information provided by the con…
Natural Language Inference (NLI) or Recognizing Textual Entailment (RTE) is the task of predicting the entailment relation between a pair of sentences (premise and hypothesis). This task has been described as a valuable testing ground for the development of semantic representations, and is a key component in natural language understanding evaluation benchmarks. Models that understand entailment should encode both, the premise and the hypothesis. However, experiments by Poliak et al. revealed a strong preference of these models towards patterns observed only in the hypothesis, based on a 10 dat…
Lexical normalisation (LN) is the process of correcting each word in a dataset to its canonical form so that it may be more easily and more accurately analysed. Most lexical normalisation systems operate at the character-level, while word-level models are seldom used. Recent language models offer solutions to the drawbacks of word-level LN models, yet, to the best of our knowledge, no research has investigated their effectiveness on LN. In this paper we introduce a word-level GRU-based LN model and investigate the effectiveness of recent embedding techniques on word-level LN. Our results show …
University of Oslo（奥斯陆大学）；University of Amsterdam（阿姆斯特丹大学）
We apply contextualised word embeddings to lexical semantic change detection in the SemEval-2020 Shared Task 1. This paper focuses on Subtask 2, ranking words by the degree of their semantic drift over time. We analyse the performance of two contextualising architectures (BERT and ELMo) and three change detection algorithms. We find that the most effective algorithms rely on the cosine similarity between averaged token embeddings and the pairwise distances between token embeddings. They outperform strong baselines by a large margin (in the post-evaluation phase, we have the best Subtask 2 subm…
Most words are ambiguous--i.e., they convey distinct meanings in different contexts--and even the meanings of unambiguous words are context-dependent. Both phenomena present a challenge for NLP. Recently, the advent of contextualized word embeddings has led to success on tasks involving lexical ambiguity, such as Word Sense Disambiguation. However, there are few tasks that directly evaluate how well these contextualized embeddings accommodate the more continuous, dynamic nature of word meaning--particularly in a way that matches human intuitions. We introduce RAW-C, a dataset of graded, human …
People convey information extremely effectively through spoken interaction using multiple channels of information transmission: the lexical channel of what is said, and the non-lexical channel of how it is said. We propose studying human perception of spoken communication as a means to better understand how information is encoded across these channels, focusing on the question 'What characteristics of communicative context affect listener's expectations of speech?'. To investigate this, we present a novel behavioural task testing whether listeners can discriminate between the true utterance in…
Yandex（扬达克斯）；University of Amsterdam（阿姆斯特丹大学）；University of Edinburgh（爱丁堡大学）；University of Zurich（苏黎世大学）
Though machine translation errors caused by the lack of context beyond one sentence have long been acknowledged, the development of context-aware NMT systems is hampered by several problems. Firstly, standard metrics are not sensitive to improvements in consistency in document-level translations. Secondly, previous work on context-aware NMT assumed that the sentence-aligned parallel data consisted of complete documents while in most practical scenarios such document-level data constitutes only a fraction of the available parallel data. To address the first issue, we perform a human study on an…
Institute for Infocomm Research, A*STAR（新加坡科技研究局资讯通信研究院）；The Chinese University of Hong Kong（香港中文大学）
Recent techniques in Question Answering (QA) have gained remarkable performance improvement with some QA models even surpassed human performance. However, the ability of these models in truly understanding the language still remains dubious and the models are revealing limitations when facing adversarial examples. To strengthen the robustness of QA models and their generalization ability, we propose a representation Enhancement via Semantic and Context constraints (ESC) approach to improve the robustness of lexical embeddings. Specifically, we insert perturbations with semantic constraints and…
State-of-the-art pretrained contextualized models (PCM) eg. BERT use tasks such as WiC and WSD to evaluate their word-in-context representations. This inherently assumes that performance in these tasks reflect how well a model represents the coupled word and context semantics. We question this assumption by presenting the first quantitative analysis on the context-word interaction being tested in major contextual lexical semantic tasks. To achieve this, we run probing baselines on masked input, and propose measures to calculate and visualize the degree of context or word biases in existing dat…
We release a new benchmark for lexical substitution, the task of finding appropriate substitutes for a target word in a context. To assist humans with writing, lexical substitution systems can suggest words that humans cannot easily think of. However, existing benchmarks depend on human recall as the only source of data, and therefore lack coverage of the substitutes that would be most helpful to humans. Furthermore, annotators often provide substitutes of low quality, which are not actually appropriate in the given context. We collect higher-coverage and higher-quality data by framing lexical…
Institute for Logic, Language and Computation（逻辑、语言与计算研究所）；University of Amsterdam（阿姆斯特丹大学）
This paper presents the first unsupervised approach to lexical semantic change that makes use of contextualised word representations. We propose a novel method that exploits the BERT neural language model to obtain representations of word usages, clusters these representations into usage types, and measures change along time with three proposed metrics. We create a new evaluation dataset and show that the model representations and the detected semantic shifts are positively correlated with human judgements. Our extensive qualitative analysis demonstrates that our method captures a variety of s…
Queen Mary University of London（伦敦玛丽女王大学）；The Alan Turing Institute（艾伦·图灵研究所）
One of the central aspects of contextualised language models is that they should be able to distinguish the meaning of lexically ambiguous words by their contexts. In this paper we investigate the extent to which the contextualised embeddings of word forms that display multiplicity of sense reflect traditional distinctions of polysemy and homonymy. To this end, we introduce an extended, human-annotated dataset of graded word sense similarity and co-predication acceptability, and evaluate how well the similarity of embeddings predicts similarity in meaning. Both types of human judgements indica…
Universitat Pompeu Fabra（庞培法布拉大学）；Idiap Research Institute（Idiap研究所）；EPFL（洛桑联邦理工学院）；ICREA（加泰罗尼亚研究与高等教育学院）；Cardiff University（卡迪夫大学）；AMPLYFI（AMPLYFI公司）
Recognizing and categorizing lexical collocations in context is useful for language learning, dictionary compilation and downstream NLP. However, it is a challenging task due to the varying degrees of frozenness lexical collocations exhibit. In this paper, we put forward a sequence tagging BERT-based model enhanced with a graph-aware transformer architecture, which we evaluate on the task of collocation recognition in context. Our results suggest that explicitly encoding syntactic dependencies in the model architecture is helpful, and provide insights on differences in collocation typification…
