The (Uncomputable!) Meaning of Ethically Charged Natural Language, for Robots, and Us, from Hypergraphical Inferential Semantics

This article discusses what kind of entity the proposed semantic Web is, principally by reference to the relationship of natural language structure to knowledge representation. The concept of the SW has two distinct origins, and this, bifurcation persists in two differing lines of SW research: one closely allied to notions of documents and natural language processing (NLP) and one not. These differences of emphasis or content carry with them different commitments about what it is to interpret a KR and what the interpretation method has to do with meaning in natural language.

Frame semantics is a theory of meaning in natural language, which defines the structure of the lexical semantic resources known as framenets. Both framenets and frame semantics have proved useful for a number of natural language processing (NLP) tasks. However, in this connection framenets have often been criticized for their limited coverage. A proposed reasonable-effort solution to this problem is to develop domain-specific (sublanguage) framenets to complement the corresponding general-language framenets for particular NLP tasks, and in the literature we find such initiatives covering domains such as medicine, soccer, and tourism. In this paper, we report on building a framenet to cover the terms and concepts encountered in descriptive linguistic grammars (written in English) i.e. a framenet for the linguistic domain (LingFN) to complement the general-language BFN.KeywordsFrame semanticsFramenetsSemantics

There has been extensive debate about the principle of compositionality among philosophers of language and formal semanticists, according to which the meaning of a complex expression is determined by the meanings of its parts and the way these parts are combined. Questions about this principle can be broadly divided into three categories: why we want compositionality, what the proper definition of compositionality should be, and how various semantic phenomena can be analyzed compositionally. In this paper, I will focus on the first question. After briefly surveying well-known arguments for compositionality, I will present two methodological arguments and examine their validity.

The paper presents a formal framework to model the fusion of gesture meaning with the meaning of the co-occurring verbal fragment. The framework is based on the formalization of two simple concepts, intersectivity and iconicity, which form the core of most descriptive accounts of the interaction of the two modalities. The formalization is presented as an extension of a well-known framework for the analysis of meaning in natural language. We claim that a proper formalization of these two concepts is sufficient to provide a principled explanation of gestures accompanying different types of linguistic expressions. The formalization also aims at providing a general mechanism (iconicity) by which the meaning of gestures is extracted from their formal appearance.

Chapter 2 - Categorial Type Logics

The project of giving an account of meaning in natural languages goes largely by assigning truth-conditional content to sentences. I will call the view that sentences have truth-conditional content propositionalism as it is common to identify the truth-conditional content of a sentence with the proposition it expresses. This content plays an important role in our explanations of the speech-acts, attitude ascriptions, and the meaning of sentences when they appear as parts of longer sentences. Much work in philosophy of language and linguistics semantics over the last halfcentury has aimed to characterize the truth-conditional content of different aspects of language.

Distributional semantics provides multidimensional, graded, empirically induced word representations that successfully capture many aspects of meaning in natural languages, as shown by a large body of research in computational linguistics; yet, its impact in theoretical linguistics has so far been limited. This review provides a critical discussion of the literature on distributional semantics, with an emphasis on methods and results that are relevant for theoretical linguistics, in three areas: semantic change, polysemy and composition, and the grammar–semantics interface (specifically, the interface of semantics with syntax and with derivational morphology). The goal of this review is to foster greater cross-fertilization of theoretical and computational approaches to language as a means to advance our collective knowledge of how it works.

Summary form only given. As the fear of semantics is declining, let's formalize the unformalizable! The address is focusing on the feasibility and necessity of accessing directly the comprehensive meaning of natural language texts, data, images, etc., to emulate human understanding by the computer. It is based on the premise that, without such understanding, no real-life application can reach the precision that human users of computational systems require. Moreover, meaning access and processing goes well beyond natural language applications, even as the scope of those keeps growing. The main obstacle for this seemingly reasonable goal is the widespread prejudice against meaning as unknowable in principle because it is “subjective,” context-sensitive, complex, messy, etc. This has not been our experience, and it is worth talking, both practically and theoretically, about the know-how that a research group should acquire to approach the processing tasks semantically. A convincing case can be made for the feasibility and affordability of the approach as well, especially in the light of our planning to make our resources Open Source, a plan that may be followed by other computational semanticists. It is hoped, therefore, that this address will further reduce the fear of, and prejudice against semantics. It will focus on our Ontological Semantic Technology, a mature formal theory of meaning that has long overcome the crippling fixation on first-order logic as the only possible formalism for everything. This fixation has prevented generations of researchers to access natural language meaning to the grain size that makes meaning representation usable, both theoretically and in practical applications. A small set of predicates using a smaller number of variables as arguments may work well for formal languages but natural language defeats such an approach. The fear of semantics is based on the faulty conclusion that, therefore, natural language meaning is not formalizable. Throw away an insufficient method, and the task becomes feasible!

We present <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">iChores</i>, an open-source, robot-agnostic interface designed to facilitate intuitive human interaction for household tasks, enabling users to direct robots through natural language and deictic gestures (pointing) to perform complex chores, such as tidying a room. The system utilizes a modular microservice architecture that integrates state-of-the-art vision, natural language understanding (NLU), and reasoning modules. We detail the implementation of the framework, provide a public repository for reproducibility, and present an evaluation of 30 scenarios of increasing complexity. The general architecture of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">iChores</i> and its various modules are briefly described. The implementation is publicly available through our GitHub project page, where all modules are provided as their own repositories.

The emergence of natural language interfaces has led to first attempts of programming in natural language. We present ProNat, a tool for script-like programming in spoken natural language (SNL). Its agent-based architecture unifies deep natural language understanding (NLU) with modular software design. ProNat focuses on the extraction of processing flows and control structures from spoken utterances. For evaluation we have begun to build a speech corpus. First experiments are conducted in the domain of domestic robotics, but ProNat's architecture makes domain acquisition easy. Test results with spoken utterances in ProNat seem promising, but much work has to be done to achieve deep NLU.

The integration of Natural Language Processing (NLP) in Human-Robot Interaction (HRI) represents a significant advancement towards achieving more natural and effective communication between humans and robots. This research explores the application of state-of-the-art NLP techniques to enhance HRI, focusing on improving robots' abilities to understand and generate human language. Key components of our approach include advanced speech recognition, natural language understanding (NLU), dialogue management, and natural language generation (NLG). We designed and implemented an HRI system that leverages models such as BERT for language understanding and GPT-3 for generating contextually appropriate responses. Our methodology involves integrating these NLP models with a robotics platform, ensuring real-time interaction capabilities while maintaining a high level of accuracy and context awareness. The system was evaluated through a series of user studies, measuring performance metrics such as accuracy, latency, and user satisfaction. Results indicate that our NLP-enhanced HRI system significantly improves the quality of interactions, demonstrating superior understanding and responsiveness compared to traditional systems. This paper discusses the implementation challenges, including computational constraints and ambiguity resolution, and provides insights into user feedback and system performance. Future work will focus on enhancing context management, exploring multimodal interaction, and addressing ethical considerations in deploying advanced HRI systems. Our findings underscore the potential of NLP to transform human-robot communication, paving the way for more intuitive and effective robotic assistants in various domains. Keywords: Human-Robot Interaction (HRI), Natural Language Processing (NLP), Conversational AI, Speech Recognition, Natural Language Understanding (NLU), Natural Language Generation (NLG), Multimodal Interaction, Dialogue Systems, Context Awareness, Emotion Recognition, Machine Learning in HRI, Personalized Interaction, User Experience (UX) in HRI, Human-Centered Design, Collaborative Robots (Cobots)

Natural language understanding (NLU) and natural language generation (NLG) are two fundamental and related tasks in building task-oriented dialogue systems with opposite objectives: NLU tackles the transformation from natural language to formal representations, whereas NLG does the reverse. A key to success in either task is parallel training data which is expensive to obtain at a large scale. In this work, we propose a generative model which couples NLU and NLG through a shared latent variable. This approach allows us to explore both spaces of natural language and formal representations, and facilitates information sharing through the latent space to eventually benefit NLU and NLG. Our model achieves state-of-the-art performance on two dialogue datasets with both flat and tree-structured formal representations. We also show that the model can be trained in a semi-supervised fashion by utilising unlabelled data to boost its performance.

Language Understanding and GenerationTrue to their name, Language-Endowed Intelligent Agents count language as a top priority.Like the humans they emulate, LEIAs learn, collaborate, and explain using language, so their language-oriented capabilities could not be more impor tant.Many of those capabilities are described in our recent book, Linguistics for the Age of AI (2021), and will not be recounted here.Instead, this chapter does this following: It provides the minimum necessary background about language understanding to make this book self-sufficient.All examples of dialog, learning, and explaining in the upcoming chapters rely on natu ral language understanding. It reports select advances in the LEIAs' language understanding module since the publication of Linguistics for the Age of AI by way of illustrating how LEIA microtheories evolve over time. It pre sents the new natu ral language generation module used by LEIAs. It juxtaposes the microtheory of construction semantics used by LEIAs with the human-oriented theory of construction grammar and shows how LEIA modeling fulfills the more rigorous demands of computation.The inevitable reliance of this chapter on material from Linguistics for the Age of AI underscores that the LEIA program of R&D can neither be encapsulated in a single monograph nor frozen at a moment in time.So, although we attempted to make this book largely free-standing, Linguistics for the Age of AI is highly recommended as a companion read. IntroductionTo serve as intelligent collaborators, LEIAs need to be able to understand and generate meaningful natu ral language.

The relationship between meaning and phonetic form in natural language is a complicated one. For the linguist doing fieldwork on an unwritten language, arriving at a sufficiently precise translation of even simple phrases is often a difficult task. Occasionally, the same grammatical feature or category (such as tense, aspect, person, number, case, etc.) will be represented by different phonetic sequences (allomorphy). The linguist is thus faced with the double problem of deciding whether different appearing forms are manifestations of the same grammatical feature and whether similar appearing forms might represent different grammatical features. It is not enough, of course, to rely totally on translated meanings since differences can be obscured quite easily by inadequate translations. We must, rather, look at glosses as but one piece of evidence in our analysis. It is the contention of this article that the determination of the phonological rules in a grammar (in the generative framework) can be a great help in determining more precisely the form and nature of the grammatical features and categories of a particular language. An excellent case demonstrating how phonological considerations elucidate grammatical features can be found in Lummi, a (Straits) Salish language formerly spoken along northern Puget Sound around Bellingham Bay in northwestern Washington State.2

Meaning and synonymy in natural languages