Natural language relies on a finite lexicon to express an unbounded set of emerging ideas. One result of this tension is the formation of new compositions, such that existing linguistic units can be combined with emerging items into novel expressions. We develop a framework that exploits the cognitive mechanisms of chaining and multimodal knowledge to predict emergent compositional expressions through time. We present the syntactic frame extension model (SFEM) that draws on the theory of chaining and knowledge from "percept", "concept", and "language" to infer how verbs extend their frames to form new compositions with existing and novel nouns. We evaluate SFEM rigorously on the 1) modalities of knowledge and 2) categorization models of chaining, in a syntactically parsed English corpus over the past 150 years. We show that multimodal SFEM predicts newly emerged verb syntax and arguments substantially better than competing models using purely linguistic or unimodal knowledge. We find support for an exemplar view of chaining as opposed to a prototype view and reveal how the joint approach of multimodal chaining may be fundamental to the creation of literal and figurative language uses including metaphor and metonymy.
Fudan University（复旦大学）；Max Planck Institute for Evolutionary Anthropology（马克斯·普朗克进化人类学研究所）
Multilingual BERT (mBERT) has demonstrated considerable cross-lingual syntactic ability, whereby it enables effective zero-shot cross-lingual transfer of syntactic knowledge. The transfer is more successful between some languages, but it is not well understood what leads to this variation and whether it fairly reflects difference between languages. In this work, we investigate the distributions of grammatical relations induced from mBERT in the context of 24 typologically different languages. We demonstrate that the distance between the distributions of different languages is highly consistent…
A range of studies have concluded that neural word prediction models can distinguish grammatical from ungrammatical sentences with high accuracy. However, these studies are based primarily on monolingual evidence from English. To investigate how these models' ability to learn syntax varies by language, we introduce CLAMS (Cross-Linguistic Assessment of Models on Syntax), a syntactic evaluation suite for monolingual and multilingual models. CLAMS includes subject-verb agreement challenge sets for English, French, German, Hebrew and Russian, generated from grammars we develop. We use CLAMS to ev…
Stockholm University（斯德哥尔摩大学）；Hebrew University of Jerusalem（耶路撒冷希伯来大学）
The patterns in which the syntax of different languages converges and diverges are often used to inform work on cross-lingual transfer. Nevertheless, little empirical work has been done on quantifying the prevalence of different syntactic divergences across language pairs. We propose a framework for extracting divergence patterns for any language pair from a parallel corpus, building on Universal Dependencies. We show that our framework provides a detailed picture of cross-language divergences, generalizes previous approaches, and lends itself to full automation. We further present a novel dat…
We are concerned with the syntactic annotation of unrestricted text. We combine a rule-based analysis with subsequent exploitation of empirical data. The rule-based surface syntactic analyser leaves some amount of ambiguity in the output that is resolved using empirical patterns. We have implemented a system for generating and applying corpus-based patterns. Some patterns describe the main constituents in the sentence and some the local context of the each syntactic function. There are several (partly) reduntant patterns, and the ``pattern'' parser selects analysis of the sentence that matches…
Guangdong University of Foreign Studies（广东外语外贸大学）；Renmin University of China（中国人民大学）；University of Montreal（蒙特利尔大学）
Deep learning models for automatic readability assessment generally discard linguistic features traditionally used in machine learning models for the task. We propose to incorporate linguistic features into neural network models by learning syntactic dense embeddings based on linguistic features. To cope with the relationships between the features, we form a correlation graph among features and use it to learn their embeddings so that similar features will be represented by similar embeddings. Experiments with six data sets of two proficiency levels demonstrate that our proposed methodology ca…
Causal Interventions Reveal Typologically Organized Syntactic Mechanisms in Multilingual Language Models
The University of Texas at Austin（德克萨斯大学奥斯汀分校）；Saarland University（萨尔大学）；Leipzig University（莱比锡大学）；Heidelberg Institute for Theoretical Studies（海德堡理论研究所）
University of Michigan（密歇根大学）；Ubiquiti, Inc.（优比快公司）
This paper presents a comparison of classification methods for linguistic typology for the purpose of expanding an extensive, but sparse language resource: the World Atlas of Language Structures (WALS) (Dryer and Haspelmath, 2013). We experimented with a variety of regression and nearest-neighbor methods for use in classification over a set of 325 languages and six syntactic rules drawn from WALS. To classify each rule, we consider the typological features of the other five rules; linguistic features extracted from a word-aligned Bible in each language; and genealogical features (genus and fam…
John Hopkins University（约翰·霍普金斯大学）；Facebook AI Research（Facebook人工智能研究院）；Catalan Institute for Research and Advanced Studies（加泰罗尼亚研究与高等研究院）；Universitat Pompeu Fabra（庞培法布拉大学）
Modern deep neural networks achieve impressive performance in engineering applications that require extensive linguistic skills, such as machine translation. This success has sparked interest in probing whether these models are inducing human-like grammatical knowledge from the raw data they are exposed to, and, consequently, whether they can shed new light on long-standing debates concerning the innate structure necessary for language acquisition. In this article, we survey representative studies of the syntactic abilities of deep networks, and discuss the broader implications that this work …
NLP has a rich history of representing our prior understanding of language in the form of graphs. Recent work on analyzing contextualized text representations has focused on hand-designed probe models to understand how and to what extent do these representations encode a particular linguistic phenomenon. However, due to the inter-dependence of various phenomena and randomness of training probe models, detecting how these representations encode the rich information in these linguistic graphs remains a challenging problem. In this paper, we propose a new information-theoretic probe, Bird's Eye, …
Shanghai Jiao Tong University（上海交通大学）；Department of Computer Science and Engineering, Shanghai Jiao Tong University（上海交通大学计算机科学与工程系）；MoE Key Lab of Artificial Intelligence, AI Institute, Shanghai Jiao Tong University（人工智能教育部重点实验室（上海交通大学人工智能研究院））
In this paper, we present a Linguistic Informed Multi-Task BERT (LIMIT-BERT) for learning language representations across multiple linguistic tasks by Multi-Task Learning (MTL). LIMIT-BERT includes five key linguistic syntax and semantics tasks: Part-Of-Speech (POS) tags, constituent and dependency syntactic parsing, span and dependency semantic role labeling (SRL). Besides, LIMIT-BERT adopts linguistics mask strategy: Syntactic and Semantic Phrase Masking which mask all of the tokens corresponding to a syntactic/semantic phrase. Different from recent Multi-Task Deep Neural Networks (MT-DNN) (…
We propose transfer learning as a method for analyzing the encoding of grammatical structure in neural language models. We train LSTMs on non-linguistic data and evaluate their performance on natural language to assess which kinds of data induce generalizable structural features that LSTMs can use for natural language. We find that training on non-linguistic data with latent structure (MIDI music or Java code) improves test performance on natural language, despite no overlap in surface form or vocabulary. To pinpoint the kinds of abstract structure that models may be encoding to lead to this i…
In this paper we identify several serious problems that arise in the use of syntactic data from the SSWL database for the purpose of computational phylogenetic reconstruction. We show that the most naive approach fails to produce reliable linguistic phylogenetic trees. We identify some of the sources of the observed problems and we discuss how they may be, at least partly, corrected by using additional information, such as prior subdivision into language families and subfamilies, and a better use of the information about ancient languages. We also describe how the use of phylogenetic algebraic…
University School for Advanced Studies IUSS Pavia（帕维亚高等研究院大学）；Ecole Polytechnique Federale de Lausanne（洛桑联邦理工学院）；University of Geneva（日内瓦大学）；Yale University（耶鲁大学）
A sharp tension exists about the nature of human language between two opposite parties: those who believe that statistical surface distributions, in particular using measures like surprisal, provide a better understanding of language processing, vs. those who believe that discrete hierarchical structures implementing linguistic information such as syntactic ones are a better tool. In this paper, we show that this dichotomy is a false one. Relying on the fact that statistical measures can be defined on the basis of either structural or non-structural models, we provide empirical evidence that o…
This research offers a new interdisciplinary approach to the field of Linguistics by using Computational Linguistics, NLP, Bayesian Statistics and Sociolinguistics methods. This thesis investigates word order change in infinitival clauses from Object-Verb (OV) to Verb-Object (VO) in the history of Latin and Old French. By applying a variationist approach, I examine a synchronic word order variation in each stage of language change, from which I infer the character, periodization and constraints of diachronic variation. I also show that in discourse-configurational languages, such as Latin and …
ECAM Rennes（雷恩ECAM学院）；Université de Bourgogne Franche-Comté（勃艮第-弗朗什-孔泰大学）；UTBM（贝尔福-蒙贝利亚尔技术大学）
This paper assumes that design language plays an important role in how designers design and on the creativity of designers. Designers use and develop models as an aid to thinking, a focus for discussion and decision-making and a means of evaluating the reliability of the proposals. This paper proposes an intelligent method for requirements engineering from natural language and their chaining toward CAD models. The transition from linguistic analysis to the representation of engineering requirements consists of the translation of the syntactic structure into semantic form represented by concept…
We describe the use of energy function optimization in very shallow syntactic parsing. The approach can use linguistic rules and corpus-based statistics, so the strengths of both linguistic and statistical approaches to NLP can be combined in a single framework. The rules are contextual constraints for resolving syntactic ambiguities expressed as alternative tags, and the statistical language model consists of corpus-based n-grams of syntactic tags. The success of the hybrid syntactic disambiguator is evaluated against a held-out benchmark corpus. Also the contributions of the linguistic and s…
Using Phylogenetic Algebraic Geometry, we analyze computationally the phylogenetic tree of subfamilies of the Indo-European language family, using data of syntactic structures. The two main sources of syntactic data are the SSWL database and Longobardi's recent data of syntactic parameters. We compute phylogenetic invariants and likelihood functions for two sets of Germanic languages, a set of Romance languages, a set of Slavic languages and a set of early Indo-European languages, and we compare the results with what is known through historical linguistics.
Recent work on evaluating grammatical knowledge in pretrained sentence encoders gives a fine-grained view of a small number of phenomena. We introduce a new analysis dataset that also has broad coverage of linguistic phenomena. We annotate the development set of the Corpus of Linguistic Acceptability (CoLA; Warstadt et al., 2018) for the presence of 13 classes of syntactic phenomena including various forms of argument alternations, movement, and modification. We use this analysis set to investigate the grammatical knowledge of three pretrained encoders: BERT (Devlin et al., 2018), GPT (Radford…
The comparative evaluation of Arabic HPSG grammar lexica requires a deep study of their linguistic coverage. The complexity of this task results mainly from the heterogeneity of the descriptive components within those lexica (underlying linguistic resources and different data categories, for example). It is therefore essential to define more homogeneous representations, which in turn will enable us to compare them and eventually merge them. In this context, we present a method for comparing HPSG lexica based on a rule system. This method is implemented within a prototype for the projection fro…
GRIMAAG；Université des Antilles et de la Guyane（安的列斯-圭亚那大学）
This article describes the design of a common syntactic description for the core grammar of a group of related dialects. The common description does not rely on an abstract sub-linguistic structure like a metagrammar: it consists in a single FS-LTAG where the actual specific language is included as one of the attributes in the set of attribute types defined for the features. When the lang attribute is instantiated, the selected subset of the grammar is equivalent to the grammar of one dialect. When it is not, we have a model of a hybrid multidialectal linguistic system. This principle is used …
Neural language models (LMs) perform well on tasks that require sensitivity to syntactic structure. Drawing on the syntactic priming paradigm from psycholinguistics, we propose a novel technique to analyze the representations that enable such success. By establishing a gradient similarity metric between structures, this technique allows us to reconstruct the organization of the LMs' syntactic representational space. We use this technique to demonstrate that LSTM LMs' representations of different types of sentences with relative clauses are organized hierarchically in a linguistically interpret…
University of Copenhagen（哥本哈根大学）；University of Edinburgh（爱丁堡大学）
Character language models have access to surface morphological patterns, but it is not clear whether or how they learn abstract morphological regularities. We instrument a character language model with several probes, finding that it can develop a specific unit to identify word boundaries and, by extension, morpheme boundaries, which allows it to capture linguistic properties and regularities of these units. Our language model proves surprisingly good at identifying the selectional restrictions of English derivational morphemes, a task that requires both morphological and syntactic awareness. …
To understand what kinds of linguistic knowledge are encoded by pretrained Chinese language models (LMs), we introduce the benchmark of Sino LINGuistics (SLING), which consists of 38K minimal sentence pairs in Mandarin Chinese grouped into 9 high-level linguistic phenomena. Each pair demonstrates the acceptability contrast of a specific syntactic or semantic phenomenon (e.g., The keys are lost vs. The keys is lost), and an LM should assign lower perplexity to the acceptable sentence. In contrast to the CLiMP dataset (Xiang et al., 2021), which also contains Chinese minimal pairs and was create…
University of Oslo（奥斯陆大学）；Uppsala University（乌普萨拉大学）；University of Copenhagen（哥本哈根大学）
Since the popularization of the Transformer as a general-purpose feature encoder for NLP, many studies have attempted to decode linguistic structure from its novel multi-head attention mechanism. However, much of such work focused almost exclusively on English -- a language with rigid word order and a lack of inflectional morphology. In this study, we present decoding experiments for multilingual BERT across 18 languages in order to test the generalizability of the claim that dependency syntax is reflected in attention patterns. We show that full trees can be decoded above baseline accuracy fr…
University of Bristol（布里斯托大学）；National Research University Higher School of Economics（高等经济学院国立研究大学）
LSTMs trained on next-word prediction can accurately perform linguistic tasks that require tracking long-distance syntactic dependencies. Notably, model accuracy approaches human performance on number agreement tasks (Gulordava et al., 2018). However, we do not have a mechanistic understanding of how LSTMs perform such linguistic tasks. Do LSTMs learn abstract grammatical rules, or do they rely on simple heuristics? Here, we test gender agreement in French which requires tracking both hierarchical syntactic structures and the inherent gender of lexical units. Our model is able to reliably pred…
Seoul National University（首尔大学）；University of Michigan（密歇根大学）
Analyzing how human beings resolve syntactic ambiguity has long been an issue of interest in the field of linguistics. It is, at the same time, one of the most challenging issues for spoken language understanding (SLU) systems as well. As syntactic ambiguity is intertwined with issues regarding prosody and semantics, the computational approach toward speech intention identification is expected to benefit from the observations of the human language processing mechanism. In this regard, we address the task with attentive recurrent neural networks that exploit acoustic and textual features simult…
Johns Hopkins University（约翰斯·霍普金斯大学）；Microsoft Research（微软研究院）；New York University（纽约大学）；Facebook AI Research（Facebook人工智能研究院）
Current language models can generate high-quality text. Are they simply copying text they have seen before, or have they learned generalizable linguistic abstractions? To tease apart these possibilities, we introduce RAVEN, a suite of analyses for assessing the novelty of generated text, focusing on sequential structure (n-grams) and syntactic structure. We apply these analyses to four neural language models (an LSTM, a Transformer, Transformer-XL, and GPT-2). For local structure - e.g., individual dependencies - model-generated text is substantially less novel than our baseline of human-gener…
University of Cambridge（剑桥大学）；ETH Zürich（苏黎世联邦理工学院）
Analysing whether neural language models encode linguistic information has become popular in NLP. One method of doing so, which is frequently cited to support the claim that models like BERT encode syntax, is called probing; probes are small supervised models trained to extract linguistic information from another model's output. If a probe is able to predict a particular structure, it is argued that the model whose output it is trained on must have implicitly learnt to encode it. However, drawing a generalisation about a model's linguistic knowledge about a specific phenomena based on what a p…
Massachusetts Institute of Technology（麻省理工学院）；Harvard University（哈佛大学）；Qatar Computing Research Institute（卡塔尔计算研究所）
Despite the recent success of deep neural networks in natural language processing (NLP), their interpretability remains a challenge. We analyze the representations learned by neural machine translation models at various levels of granularity and evaluate their quality through relevant extrinsic properties. In particular, we seek answers to the following questions: (i) How accurately is word-structure captured within the learned representations, an important aspect in translating morphologically-rich languages? (ii) Do the representations capture long-range dependencies, and effectively handle …
UC Irvine（加州大学欧文分校）；Harvard University（哈佛大学）；Primate Research Institute, Kyoto University（京都大学灵长类研究所）；MIT（麻省理工学院）
Recurrent neural networks (RNNs) are the state of the art in sequence modeling for natural language. However, it remains poorly understood what grammatical characteristics of natural language they implicitly learn and represent as a consequence of optimizing the language modeling objective. Here we deploy the methods of controlled psycholinguistic experimentation to shed light on to what extent RNN behavior reflects incremental syntactic state and grammatical dependency representations known to characterize human linguistic behavior. We broadly test two publicly available long short-term memor…
A standard approach to evaluating language models analyzes how models assign probabilities to valid versus invalid syntactic constructions (i.e. is a grammatical sentence more probable than an ungrammatical sentence). Our work uses ambiguous relative clause attachment to extend such evaluations to cases of multiple simultaneous valid interpretations, where stark grammaticality differences are absent. We compare model performance in English and Spanish to show that non-linguistic biases in RNN LMs advantageously overlap with syntactic structure in English but not Spanish. Thus, English models m…
It is reasonable to hypothesize that the divergence patterns formulated by historical linguists and typologists reflect constraints on human languages, and are thus consistent with Second Language Acquisition (SLA) in a certain way. In this paper, we validate this hypothesis on ten Indo-European languages. We formalize the delexicalized transfer as interpretable tree-to-string and tree-to-tree patterns which can be automatically induced from web data by applying neural syntactic parsing and grammar induction technologies. This allows us to quantitatively probe cross-linguistic transfer and ext…
There is an increasing amount of evidence that in cases with little or no data in a target language, training on a different language can yield surprisingly good results. However, currently there are no established guidelines for choosing the training (source) language. In attempt to solve this issue we thoroughly analyze a state-of-the-art multilingual model and try to determine what impacts good transfer between languages. As opposed to the majority of multilingual NLP literature, we don't only train on English, but on a group of almost 30 languages. We show that looking at particular syntac…
