Compositionality is a widely discussed property of natural languages, although its exact definition has been elusive. We focus on the proposal that compositionality can be assessed by measuring meaning-form correlation. We analyze meaning-form correlation on three sets of languages: (i) artificial toy languages tailored to be compositional, (ii) a set of English dictionary definitions, and (iii) a set of English sentences drawn from literature. We find that linguistic phenomena such as synonymy and ungrounded stop-words weigh on MFC measurements, and that straightforward methods to mitigate their effects have widely varying results depending on the dataset they are applied to. Data and code are made publicly available.
To understand a sentence like "whereas only 10% of White Americans live at or below the poverty line, 28% of African Americans do" it is important not only to identify individual facts, e.g., poverty rates of distinct demographic groups, but also the higher-order relations between them, e.g., the disparity between them. In this paper, we propose the task of Textual Analogy Parsing (TAP) to model this higher-order meaning. The output of TAP is a frame-style meaning representation which explicitly specifies what is shared (e.g., poverty rates) and what is compared (e.g., White Americans vs. Afri…
University of California, Los Angeles（加利福尼亚大学洛杉矶分校）
Measuring meaning is a central problem in cultural sociology and word embeddings may offer powerful new tools to do so. But like any tool, they build on and exert theoretical assumptions. In this paper I theorize the ways in which word embeddings model three core premises of a structural linguistic theory of meaning: that meaning is relational, coherent, and may be analyzed as a static system. In certain ways, word embedding methods are vulnerable to the same, enduring critiques of these premises. In other ways, they offer novel solutions to these critiques. More broadly, formalizing the study…
Allen Institute for AI（艾伦人工智能研究所）；Bar Ilan University（巴伊兰大学）；Hebrew University of Jerusalem（耶路撒冷希伯来大学）；University of Washington（华盛顿大学）
Language models trained on billions of tokens have recently led to unprecedented results on many NLP tasks. This success raises the question of whether, in principle, a system can ever ``understand'' raw text without access to some form of grounding. We formally investigate the abilities of ungrounded systems to acquire meaning. Our analysis focuses on the role of ``assertions'': textual contexts that provide indirect clues about the underlying semantics. We study whether assertions enable a system to emulate representations preserving semantic relations like equivalence. We find that assertio…
Over the past 50 years many have debated what representation should be used to capture the meaning of natural language utterances. Recently new needs of such representations have been raised in research. Here I survey some of the interesting representations suggested to answer for these new needs.
Given a specific discourse, which discourse properties trigger the use of metaphorical language, rather than using literal alternatives? For example, what drives people to say "grasp the meaning" rather than "understand the meaning" within a specific context? Many NLP approaches to metaphorical language rely on cognitive and (psycho-)linguistic insights and have successfully defined models of discourse coherence, abstractness and affect. In this work, we build five simple models relying on established cognitive and linguistic properties -- frequency, abstractness, affect, discourse coherence a…
The main goal of this master's thesis is to introduce Quantum Natural Language Processing (QNLP) in a way understandable by both the NLP engineer and the quantum computing practitioner. QNLP is a recent application of quantum computing that aims at representing sentences' meaning as vectors encoded into quantum computers. To achieve this, the distributional meaning of words is extended by the compositional meaning of sentences (DisCoCat model) : the vectors representing words' meanings are composed through the syntactic structure of the sentence. This is done using an algorithm based on tensor…
In this workshop paper, we use an empirical example from our ongoing fieldwork, to showcase the complexity and situatedness of the process of making sense of algorithmic results; i.e. how to evaluate, validate, and contextualize algorithmic outputs. So far, in our research work, we have focused on such sense-making processes in data analytic learning environments such as classrooms and training workshops. Multiple moments in our fieldwork suggest that meaning, in data analytics, is constructed through an iterative and reflexive dialogue between data, code, assumptions, prior knowledge, and alg…
To What Extent Do Large Language Models Understand Bangla Idioms?
Technical University Dortmund（多特蒙德工业大学）；Research Center Trustworthy Data Science and Security（可信数据科学与安全研究中心）；BRAC University（BRAC大学）；United International University（联合国际大学）
What's the Meaning of Superhuman Performance in Today's NLU?
Babelscape（巴别斯克）；Sapienza University of Rome（罗马第一大学）；University of Groningen（格罗宁根大学）；German Research Center for AI (DFKI)（德国人工智能研究中心）；Charles University（查理大学）；University of Copenhagen（哥本哈根大学）；University of Melbourne（墨尔本大学）；Carnegie Mellon University（卡内基梅隆大学）；Saarland University（萨尔大学）；Jožef Stefan Institute（约泽夫·斯特凡研究所）；University of Ljubljana（卢布尔雅那大学）；Cardiff University（卡迪夫大学）；University of Zurich（苏黎世大学）；University of Edinburgh（爱丁堡大学）；University of Amsterdam（阿姆斯特丹大学）
Neural language models exhibit impressive performance on a variety of tasks, but their internal reasoning may be difficult to understand. Prior art aims to uncover meaningful properties within model representations via probes, but it is unclear how faithfully such probes portray information that the models actually use. To overcome such limitations, we propose a technique, inspired by causal analysis, for generating counterfactual embeddings within models. In experiments testing our technique, we produce evidence that suggests some BERT-based models use a tree-distance-like representation of s…
Recent work in NLP shows that LSTM language models capture compositional structure in language data. For a closer look at how these representations are composed hierarchically, we present a novel measure of interdependence between word meanings in an LSTM, based on their interactions at the internal gates. To explore how compositional representations arise over training, we conduct simple experiments on synthetic data, which illustrate our measure by showing how high interdependence can hurt generalization. These synthetic experiments also illustrate a specific hypothesis about how hierarchica…
MIT（麻省理工学院）；New York University（纽约大学）；Allen Institute for Artificial Intelligence（艾伦人工智能研究所）；University of Washington（华盛顿大学）
Many current NLP systems are built from language models trained to optimize unsupervised objectives on large amounts of raw text. Under what conditions might such a procedure acquire meaning? Our systematic experiments with synthetic data reveal that, with languages where all expressions have context-independent denotations (i.e., languages with strong transparency), both autoregressive and masked language models successfully learn to emulate semantic relations between expressions. However, when denotations are changed to be context-dependent with the language otherwise unmodified, this abilit…
University of California, Berkeley（加州大学伯克利分校）；DeepMind（深度思维）
The widespread success of large language models (LLMs) has been met with skepticism that they possess anything like human concepts or meanings. Contrary to claims that LLMs possess no meaning whatsoever, we argue that they likely capture important aspects of meaning, and moreover work in a way that approximates a compelling account of human cognition in which meaning arises from conceptual role. Because conceptual role is defined by the relationships between internal representational states, meaning cannot be determined from a model's architecture, training data, or objective function, but onl…
Recent work in NLP shows that LSTM language models capture hierarchical structure in language data. In contrast to existing work, we consider the \textit{learning} process that leads to their compositional behavior. For a closer look at how an LSTM's sequential representations are composed hierarchically, we present a related measure of Decompositional Interdependence (DI) between word meanings in an LSTM, based on their gate interactions. We connect this measure to syntax with experiments on English language data, where DI is higher on pairs of words with lower syntactic distance. To explore …
Columbia University（哥伦比亚大学）；École Polytechnique Fédérale de Lausanne（洛桑联邦理工学院）；Carnegie Observatories（卡内基天文台）；University of Southern California（南加州大学）；Sapienza University of Rome（罗马第一大学）；Stanford University（斯坦福大学）；University of California, Berkeley（加州大学伯克利分校）；Haas Business School（哈斯商学院）
Natural language processing models learn word representations based on the distributional hypothesis, which asserts that word context (e.g., co-occurrence) correlates with meaning. We propose that $n$-grams composed of random character sequences, or $garble$, provide a novel context for studying word meaning both within and beyond extant language. In particular, randomly generated character $n$-grams lack meaning but contain primitive information based on the distribution of characters they contain. By studying the embeddings of a large corpus of garble, extant language, and pseudowords using …
As language models such as GPT-3 become increasingly successful at generating realistic text, questions about what purely text-based modeling can learn about the world have become more urgent. Is text purely syntactic, as skeptics argue? Or does it in fact contain some semantic information that a sufficiently sophisticated language model could use to learn about the world without any additional inputs? This paper describes a new model that suggests some qualified answers to those questions. By theorizing the relationship between text and the world it describes as an equilibrium relationship be…
University of Michigan（密歇根大学）；Westlake University（西湖大学）
Text-to-SQL has attracted attention from both the natural language processing and database communities because of its ability to convert the semantics in natural language into SQL queries and its practical application in building natural language interfaces to database systems. The major challenges in text-to-SQL lie in encoding the meaning of natural utterances, decoding to SQL queries, and translating the semantics between these two forms. These challenges have been addressed to different extents by the recent advances. However, there is still a lack of comprehensive surveys for this task. T…
CoLabPotsdam（波茨坦合作实验室）；University of Potsdam（波茨坦大学）
The striking recent advances in eliciting seemingly meaningful language behaviour from language-only machine learning models have only made more apparent, through the surfacing of clear limitations, the need to go beyond the language-only mode and to ground these models "in the world". Proposals for doing so vary in the details, but what unites them is that the solution is sought in the addition of non-linguistic data types such as images or video streams, while largely keeping the mode of learning constant. I propose a different, and more wide-ranging conception of how grounding should be und…
