Large Language Models (LLMs) have been transformative. They are pre-trained foundational models that are self-supervised and can be adapted with fine tuning to a wide range of natural language tasks, each of which previously would have required a separate network model. This is one step closer to the extraordinary versatility of human language. GPT-3 and more recently LaMDA can carry on dialogs with humans on many topics after minimal priming with a few examples. However, there (...) has been a wide range of reactions and debate on whether these LLMs understand what they are saying or exhibit signs of intelligence. This high variance is exhibited in three interviews with LLMs reaching wildly different conclusions. A new possibility was uncovered that could explain this divergence. What appears to be intelligence in LLMs may in fact be a mirror that reflects the intelligence of the interviewer, a remarkable twist that could be considered a Reverse Turing Test. If so, then by studying interviews we may be learning more about the intelligence and beliefs of the interviewer than the intelligence of the LLMs. As LLMs become more capable they may transform the way we interact with machines and how they interact with each other. Increasingly, LLMs are being coupled with sensorimotor devices. LLMs can talk the talk, but can they walk the walk? A road map for achieving artificial general autonomy is outlined with seven major improvements inspired by brain systems. LLMs could be used to uncover new insights into brain function by downloading brain data during natural behaviors. (shrink)

To what degree can language be acquired from linguistic input alone? This question has vexed scholars for millennia and is still a major focus of debate in the cognitive science of language. The complexity of human language has hampered progress because studies of language–especially those involving computational modeling–have only been able to deal with small fragments of our linguistic skills. We suggest that the most recent generation of Large Language Models (LLMs) might finally (...) provide the computational tools to determine empirically how much of the human language ability can be acquired from linguistic experience. LLMs are sophisticated deep learning architectures trained on vast amounts of natural language data, enabling them to perform an impressive range of linguistic tasks. We argue that, despite their clear semantic and pragmatic limitations, LLMs have already demonstrated that human-like grammatical language can be acquired without the need for a built-in grammar. Thus, while there is still much to learn about how humans acquire and use language, LLMs provide full-fledged computational models for cognitive scientists to empirically evaluate just how far statistical learning might take us in explaining the full complexity of human language. (shrink)

Sallie McFague’s theological models construct a tensive relationship between conceptual structures and symbolic, metaphorical language to interpret the defining and elusive aspects of theological phenomena and loci. Computational models of language can extend and formalize the conceptual structures of theological models to develop computer-augmented interpretations of theological texts. Previously unclear is whether computational models can retain the tensive symbolism essential for theological investigation. I demonstrate affirmatively by constructing a computational topic model (...) of the moral theology of Thomas Aquinas from Summa Theologica (Second Part, in English translation) useful for interpreting not only the Thomistic text but also recent papal encyclicals. (shrink)

Are LLMs cultural technologies like photocopiers or printing presses, which transmit information but cannot create new content? A challenge for this idea, which we call bibliotechnism, is that LLMs generate novel text. We begin with a defense of bibliotechnism, showing how even novel text may inherit its meaning from original human-generated text. We then argue that bibliotechnism faces an independent challenge from examples in which LLMs generate novel reference, using new names to refer to new entities. Such examples could be (...) explained if LLMs were not cultural technologies but had beliefs, desires, and intentions. According to interpretationism in the philosophy of mind, a system has such attitudes if and only if its behavior is well explained by the hypothesis that it does. Interpretationists may hold that LLMs have attitudes, and thus have a simple solution to the novel reference problem. We emphasize, however, that interpretationism is compatible with very simple creatures having attitudes and differs sharply from views that presuppose these attitudes require onsciousness, sentience, or intelligence (topics about which we make no claims). (shrink)

This letter explores the intricate historical and contemporary links between large language models (LLMs) and cognitive science through the lens of information theory, statistical language models, and socioanthropological linguistic theories. The emergence of LLMs highlights the enduring significance of information‐based and statistical learning theories in understanding human communication. These theories, initially proposed in the mid‐20th century, offered a visionary framework for integrating computational science, social sciences, and humanities, which nonetheless was not fully fulfilled at that (...) time. The subsequent development of sociolinguistics and linguistic anthropology, especially since the 1970s, provided critical perspectives and empirical methods that both challenged and enriched this framework. This letter proposes that two pivotal concepts derived from this development, metapragmatic function and indexicality, offer a fruitful theoretical perspective for integrating the semantic, textual, and pragmatic, contextual dimensions of communication, an amalgamation that contemporary LLMs have yet to fully achieve. The author believes that contemporary cognitive science is at a crucial crossroads, where fostering interdisciplinary dialogues among computational linguistics, social linguistics and linguistic anthropology, and cognitive and social psychology is in particular imperative. Such collaboration is vital to bridge the computational, cognitive, and sociocultural aspects of human communication and human−AI interaction, especially in the era of large language and multimodal models and human‐centric Artificial Intelligence (AI). (shrink)

In natural language understanding, a crucial goal is correctly interpreting open-textured phrases. In practice, disagreements over the meanings of open-textured phrases are often resolved through the generation and evaluation of interpretive arguments, arguments designed to support or attack a specific interpretation of an expression within a document. In this paper, we discuss some of our work towards the goal of automatically generating and evaluating interpretive arguments. We have curated a set of rules from the code of ethics of various (...) professional organizations and a set of associated scenarios that are ambiguous with respect to some open-textured phrase within the rule. We collected and evaluated arguments from both human annotators and state-of-the-art generative language models in order to determine the relative quality and persuasiveness of both sets of arguments. Finally, we performed a Turing test-inspired study in order to assess whether human annotators can tell the difference between human arguments and machine-generated arguments. The results show that machine-generated arguments, when prompted a certain way, can be consistently rated as more convincing than human-generated arguments, and to the untrained eye, the machine-generated arguments can convincingly sound human-like. (shrink)

Large Language Models (LLMs) are generative AI systems capable of producing original texts based on inputs about topic and style provided in the form of prompts or questions. The introduction of the outputs of these systems into human discursive practices poses unprecedented moral and political questions. The article articulates an analysis of the moral status of these systems and their interactions with human interlocutors based on the Habermasian theory of communicative action. The analysis explores, among other things, Habermas's (...) inquiries into the analogy between human minds and computers, and into the status of atypical participants in the linguistic community such as genetically modified subjects and animals. Major conclusions are the LLMs seem to qualify as authors that originally participate in discursive practices but do display only a structurally derivative form of communicative competence and fail to meet the status of communicative agents. In this sense, while the contribution of AI writing systems in public discourse and deliberation can support the process of mutual understanding within the community of speakers, the human actors involved in the development, use, and diffusion of these systems share a collective responsibility for the disclosure of AI authorship and verification and adjudication of validity claims. (shrink)

This book compiles and reviews the most prominent linguistic theories into a single source that serves as an essential reference for future solutions to one of the most important challenges of our age.

Large Language Models (LLMs) could be a useful tool for lawyers. However, empirical research on their effectiveness in conducting legal tasks is scant. We study securities cases involving cryptocurrencies as one of numerous contexts where AI could support the legal process, studying GPT-3.5’s legal reasoning and ChatGPT’s legal drafting capabilities. We examine whether a) GPT-3.5 can accurately determine which laws are potentially being violated from a fact pattern, and b) whether there is a difference in juror decision-making based (...) on complaints written by a lawyer compared to ChatGPT. We feed fact patterns from real-life cases to GPT-3.5 and evaluate its ability to determine correct potential violations from the scenario and exclude spurious violations. Second, we had mock jurors assess complaints written by ChatGPT and lawyers. GPT-3.5’s legal reasoning skills proved weak, though we expect improvement in future models, particularly given the violations it suggested tended to be correct (it merely missed additional, correct violations). ChatGPT performed better at legal drafting, and jurors’ decisions were not statistically significantly associated with the author of the document upon which they based their decisions. Because GPT-3.5 cannot satisfactorily conduct legal reasoning tasks, it would be unlikely to be able to help lawyers in a meaningful way at this stage. However, ChatGPT’s drafting skills (though, perhaps, still inferior to lawyers) could assist lawyers in providing legal services. Our research is the first to systematically study an LLM’s legal drafting and reasoning capabilities in litigation, as well as in securities law and cryptocurrency-related misconduct. (shrink)

The surge in interest in natural language processing in artificial intelligence has led to an explosion of new language models capable of engaging in plausible language use. But ensuring these language models produce honest, helpful, and inoffensive outputs has proved difficult. In this paper, I argue problems of inappropriate content in current, autoregressive language models—such as ChatGPT and Gemini—are inescapable; merely predicting the next word is incompatible with reliably providing appropriate outputs. The (...) various fine-tuning methods, while helpful, cannot transform the model from mere next word prediction to the kind of planning and forethought necessary for saying the right thing. The upshot is that these models will increasingly churn out bland, generic responses that will still fail to be accurate or appropriate. (shrink)

This paper analyses the phenomenology and epistemology of chatbots such as ChatGPT and Bard. The computational architecture underpinning these chatbots are large language models (LLMs), which are generative AI (Artificial Intelligence) systems trained on a massive dataset of text extracted from the Web. We conceptualise these LLMs as multifunctional computational cognitive artifacts, used for various cognitive tasks such as translating, summarizing, answering questions, information-seeking, and much more. Phenomenologically, LLMs can be experienced as a “quasi-other”; when that (...) happens, users anthropomorphise them. For most users, current LLMs are black boxes, i.e., for the most part, they lack data transparency and algorithmic transparency. They can, however, be phenomenologically and informationally transparent, in which case there is an interactional flow. Anthropomorphising and interactional flow can, in some users, create an attitude of (unwarranted) trust towards the output LLMs generate. We conclude this paper by drawing on the epistemology of trust and testimony to examine the epistemic implications of these dimensions. Whilst LLMs generally generate accurate responses, we observe two epistemic pitfalls. Ideally, users should be able to match the level of trust that they place in LLMs to the degree that LLMs are trustworthy. However, both their data and algorithmic opacity and their phenomenological and informational transparency can make it difficult for users to calibrate their trust correctly. The effects of these limitations are twofold: users may adopt unwarranted attitudes of trust towards the outputs of LLMs (which is particularly problematic when LLMs hallucinate), and the trustworthiness of LLMs may be undermined. (shrink)

The psycholinguistic literature has identified two syntactic adaptation effects in language production: rapidly decaying short-term priming and long-lasting adaptation. To explain both effects, we present an ACT-R model of syntactic priming based on a wide-coverage, lexicalized syntactic theory that explains priming as facilitation of lexical access. In this model, two well-established ACT-R mechanisms, base-level learning and spreading activation, account for long-term adaptation and short-term priming, respectively. Our model simulates incremental language production and in a series of modeling studies, (...) we show that it accounts for (a) the inverse frequency interaction; (b) the absence of a decay in long-term priming; and (c) the cumulativity of long-term adaptation. The model also explains the lexical boost effect and the fact that it only applies to short-term priming. We also present corpus data that verify a prediction of the model, that is, that the lexical boost affects all lexical material, rather than just heads. (shrink)

Named entity recognition (NER) is a very relevant task for text information retrieval in natural language processing (NLP) problems. Most recent state-of-the-art NER methods require humans to annotate and provide useful data for model training. However, using human power to identify, circumscribe and label entities manually can be very expensive in terms of time, money, and effort. This paper investigates the use of prompt-based language models (OpenAI’s GPT-3) and weak supervision in the legal domain. We apply both (...) strategies as alternative approaches to the traditional human-based annotation method, relying on computer power instead human effort for labeling, and subsequently compare model performance between computer and human-generated data. We also introduce combinations of all three mentioned methods (prompt-based, weak supervision, and human annotation), aiming to find ways to maintain high model efficiency and low annotation costs. We showed that, despite human labeling still maintaining better overall performance results, the alternative strategies and their combinations presented themselves as valid options, displaying positive results and similar model scores at lower costs. Final results demonstrate preservation of human-trained models scores averaging 74.0% for GPT-3, 95.6% for weak supervision, 90.7% for GPT + weak supervision combination, and 83.9% for GPT + 30% human-labeling combination. (shrink)

The use of language models in Web applications and other areas of computing and business have grown significantly over the last five years. One reason for this growth is the improvement in performance of language models on a number of benchmarks — but a side effect of these advances has been the adoption of a “bigger is always better” paradigm when it comes to the size of training, testing, and challenge datasets. Drawing on previous criticisms of (...) this paradigm as applied to large training datasets crawled from pre-existing text on the Web, we extend the critique to challenge datasets custom-created by crowdworkers. We present several sets of criticisms, where ethical and scientific issues in language model research reinforce each other: labour injustices in crowdwork, dataset quality and inscrutability, inequities in the research community, and centralized corporate control of the technology. We also present a new type of tool for researchers to use in examining large datasets when evaluating them for quality. (shrink)

The article explores the use of the “ChatGPT” artificial intelligence language model in the Human Sciences field. ChatGPT uses natural language processing techniques to imitate human language and engage in artificial conversations. While the platform has gained attention from the scientific community, opinions on its usage are divided. The article presents some conversations with ChatGPT to examine ethical, relational and linguistic issues related to human-computer interaction (HCI) and assess its potential for Human Sciences research. The interaction with (...) the platform recalls the “uncanny valley” phenomenon still known to Social Robotics. While ChatGPT can be beneficial, it requires proper supervision and verification of its results. Furthermore, new research methods must be developed for qualitative, quantitative, and mixed methods. (shrink)

This paper explores the challenges posed by the rapid advancement of artificial intelligence specifically Large Language Models (LLMs). I show that traditional linguistic theories and corpus studies are being outpaced by LLMs’ computational sophistication and low perplexity levels. In order to address these challenges, I suggest a focus on language as a cognitive tool shaped by embodied-environmental imperatives in the context of Agentive Cognitive Construction Grammar. To that end, I introduce an Embodied Human Language Model (...) (EHLM), inspired by Active Inference research, as a promising alternative that integrates sensory input, embodied representations, and adaptive strategies for contextualized analysis and conceptual utility maximization. By incorporating Active Inference, which sees perception as inferring the world's state from sensory data, the findings reveal that the characterization of the English modal be able to, as a triadic construction encoding biological intelligent agency, introduces a more plausible theoretical basis for the positing of linguistic constructions. This emphasizes the crucial role of embodied human language models in the comprehension of how humans construct preferred futures through language. (shrink)

This unique book presents a comprehensive and rigorous treatment of the theory of computability which is introductory yet self-contained. It takes a novel approach by looking at the subject using computation models rather than a limitation orientation, and is the first book of its kind to include software. Accompanying software simulations of almost all computational models are available for use in conjunction with the text, and numerous examples are provided on disk in a user-friendly format. Its applications (...) to computer science itself include interesting links to programming language theory, compiler design theory, and algorithm design. The software, numerous examples, and solutions make this book ideal for self-study by computer scientists and mathematicians alike. (shrink)

Recent artificial intelligence advances, especially those of large language models (LLMs), have increasingly shown glimpses of human-like intelligence. This has led to bold claims that these systems are no longer a mere “it” but now a “who,” a kind of person deserving respect. In this paper, I argue that this view depends on a Cartesian account of personhood, on which identifying someone as a person is based on their cognitive sophistication and ability to address common-sense reasoning problems. I (...) contrast this with a different account of personhood, one where an agent is a person if they are autonomous, responsive to norms, and culpable for their actions. On this latter account, I show that LLMs are not person-like, as evidenced by their propensity for dishonesty, inconsistency, and offensiveness. Moreover, I argue current LLMs, given the way they are designed and trained, cannot be persons—either social or Cartesian. The upshot is that contemporary LLMs are not, and never will be, persons. (shrink)

In artificial intelligence literature, “delusions” are characterized as the generation of unfaithful output from reliable source content. There is an extensive literature on computer-generated delusions, ranging from visual hallucinations, like the production of nonsensical images in Computer Vision, to nonsensical text generated by (natural) language models, but this literature is predominantly taxonomic. In a recent research paper, however, a group of scientists from DeepMind successfully presented a formal treatment of an entire class of delusions in generative AI (...) class='Hi'>models (i.e., models based on a transformer architecture, both with and without RLHF—reinforcement learning with human feedback, such as BERT, GPT-3 or the more recent GPT-3.5), referred to as auto-suggestive delusions. Auto-suggestive delusions are not mere unfaithful output, but are self-induced by the transformer models themselves. Typically, these delusions have been subsumed under the concept of exposure bias, but exposure bias alone does not elucidate their nature. In order to address their nature, I will introduce a formal framework that clarifies the probabilistic delusions capable of explaining exposure bias in a broad manner. This will serve as the foundation for exploring auto-suggestive delusions in language models. Next, an examination of self- or auto-suggestive delusions will be undertaken, by drawing an analogy with the rule-following problematic from the philosophy of mind and language. Finally, I will argue that this comprehensive approach leads to the suggestion that transformers, large language models in particular, may develop in a manner that touches upon solipsism and the emergence of a private language, in a weak sense. (shrink)

What makes a word easy to learn? Early‐learned words are frequent and tend to name concrete referents. But words typically do not occur in isolation. Some words are predictable from their contexts; others are less so. Here, we investigate whether predictability relates to when children start producing different words (age of acquisition; AoA). We operationalized predictability in terms of a word's surprisal in child‐directed speech, computed using n‐gram and long‐short‐term‐memory (LSTM) language models. Predictability derived from LSTMs was generally (...) a better predictor than predictability derived from n‐gram models. Across five languages, average surprisal was positively correlated with the AoA of predicates and function words but not nouns. Controlling for concreteness and word frequency, more predictable predicates and function words were learned earlier. Differences in predictability between languages were associated with cross‐linguistic differences in AoA: the same word (when it was a predicate) was produced earlier in languages where the word was more predictable. (shrink)

This research attempts to understand howchildren learn to use language. Instead ofusing syntax-based grammar rules to model thedifferences between children''s language andadult language, as has been done in the past, anew model is proposed. In the new researchmodel, children acquire language by listeningto the examples of speech that they hear intheir environment and subsequently use thespeech examples that have been previously heardin similar contextual situations. A computermodel is generated to simulate this new modelof language acquisition. (...) The MALL computerprogram will listen to examples of humanspeech, as would occur around a child, and thentry to use these examples in new situationsthat are similar to the contextual situationsin which the language examples were heard. This will provide a better understanding of howchildren learn to use language and howeducators can assist or improve the languagelearning process by providing required examplesof speech or by helping children to develop abetter understanding of similarities betweenvarious contexts. (shrink)

Several evolutionary accounts of human social cognition posit that language has co-evolved with the sophisticated mindreading abilities of modern humans. It has also been argued that these mindreading abilities are the product of cultural, rather than biological, evolution. Taken together, these claims suggest that the evolution of language has played an important role in the cultural evolution of human social cognition. Here we present a new computational model which formalises the assumptions that underlie this hypothesis, in order (...) to explore how language and mindreading interact through cultural evolution. This model treats communicative behaviour as an interplay between the context in which communication occurs, an agent’s individual perspective on the world, and the agent’s lexicon. However, each agent’s perspective and lexicon are private mental representations, not directly observable to other agents. Learners are therefore confronted with the task of jointly inferring the lexicon and perspective of their cultural parent, based on their utterances in context. Simulation results show that given these assumptions, an informative lexicon evolves not just under a pressure to be successful at communicating, but also under a pressure for accurate perspective-inference. When such a lexicon evolves, agents become better at inferring others’ perspectives; not because their innate ability to learn about perspectives changes, but because sharing a language with others helps them to do so. (shrink)

Recent advancements in large language models (LLMs) have ushered in a transformative phase in artificial intelligence (AI). Unlike conventional AI, LLMs excel in facilitating fluid human–computer d...

In this research, patent prosecution is conceptualized as a system of reinforcement learning from human feedback. The objective of the system is to increase the likelihood for a language model to generate patent claims that have a higher chance of being granted. To showcase the controllability of the language model, the system learns from granted patents and pre-grant applications with different rewards. The status of “granted” and “pre-grant” are perceived as labeled human feedback implicitly. In addition, specific to (...) patent drafting, the experiments in this research demonstrate the model’s capability to learn from adjusting claim length and inclusion of limiting terms for narrowing claim scope. As proof of concept, the experiments focus on claim ones only and the training data originates from a patent dataset tailored specifically for artificial intelligence. Although the available human feedback in patent prosecution are limited and the quality of generated patent text requires improvement, the experiments following the 3-stage reinforcement learning from human feedback have demonstrated that generative language models are capable of reflecting the human feedback or intent in patent prosecution. To enhance the usability of language models, the implementation in this research utilizes modern techniques that enable execution on a single consumer-grade GPU. The demonstrated proof of concept, which reduces hardware requirements, will prove valuable in the future as more human feedback in patent prosecution become available for broader use, either within patent offices or in the public domain. (shrink)

There is substantial variation in language experience, yet there is surprising similarity in the language structure acquired. Constraints on language structure may be external modulators that result in this canalization of language structure, or else they may derive from the broader, communicative environment in which language is acquired. In this paper, the latter perspective is tested for its adequacy in explaining robustness of language learning to environmental variation. A computational model of word learning (...) from cross-situational, multimodal information was constructed and tested. Key to the model's robustness was the presence of multiple, individually unreliable information sources to support learning. This “degeneracy” in the language system has a detrimental effect on learning, compared to a noise-free environment, but has a critically important effect on acquisition of a canalized system that is resistant to environmental noise in communication. (shrink)

Available corpora for Argument Mining differ along several axes, and one of the key differences is the presence (or absence) of discourse markers to signal argumentative content. Exploring effective ways to use discourse markers has received wide attention in various discourse parsing tasks, from which it is well-known that discourse markers are strong indicators of discourse relations. To improve the robustness of Argument Mining systems across different genres, we propose to automatically augment a given text with discourse markers such that (...) all relations are explicitly signaled. Our analysis unveils that popular language models taken out-of-the-box fail on this task; however, when fine-tuned on a new heterogeneous dataset that we construct (including synthetic and real examples), they perform considerably better. We demonstrate the impact of our approach on an Argument Mining downstream task, evaluated on different corpora, showing that language models can be trained to automatically fill in discourse markers across different corpora, improving the performance of a downstream model in some, but not all, cases. Our proposed approach can further be employed as an assistant tool for better discourse understanding. (shrink)

There is substantial variation in language experience, yet there is surprising similarity in the language structure acquired. Constraints on language structure may be external modulators that result in this canalization of language structure, or else they may derive from the broader, communicative environment in which language is acquired. In this paper, the latter perspective is tested for its adequacy in explaining robustness of language learning to environmental variation. A computational model of word learning (...) from cross-situational, multimodal information was constructed and tested. Key to the model's robustness was the presence of multiple, individually unreliable information sources to support learning. This “degeneracy” in the language system has a detrimental effect on learning, compared to a noise-free environment, but has a critically important effect on acquisition of a canalized system that is resistant to environmental noise in communication. (shrink)

This article contributes to the debate around the abilities of large language models such as GPT-3, dealing with: firstly, evaluating how well GPT does in the Turing Test, secondly the limits of such models, especially their tendency to generate falsehoods, and thirdly the social consequences of the problems these models have with truth-telling. We start by formalising the recently proposed notion of reversible questions, which Floridi & Chiriatti propose allow one to ‘identify the nature of the (...) source of their answers’, as a probabilistic measure based on Item Response Theory from psychometrics. Following a critical assessment of the methodology which led previous scholars to dismiss GPT’s abilities, we argue against claims that GPT-3 completely lacks semantic ability. Using ideas of compression, priming, distributional semantics and semantic webs we offer our own theory of the limits of large language models like GPT-3, and argue that GPT can competently engage in various semantic tasks. The real reason GPT’s answers seem senseless being that truth-telling is not amongst them. We claim that these kinds of models cannot be forced into producing only true continuation, but rather to maximise their objective function they strategize to be plausible instead of truthful. This, we moreover claim, can hijack our intuitive capacity to evaluate the accuracy of its outputs. Finally, we show how this analysis predicts that a widespread adoption of language generators as tools for writing could result in permanent pollution of our informational ecosystem with massive amounts of very plausible but often untrue texts. (shrink)

What is the mind? How does it work? How does it influence behavior? Some psychologists hope to answer such questions in terms of concepts drawn from computer science and artificial intelligence. They test their theories by modeling mental processes in computers. This book shows how computer models are used to study many psychological phenomena--including vision, language, reasoning, and learning. It also shows that computer modeling involves differing theoretical approaches. Computational psychologists disagree about some basic questions. For instance, (...) should the mind be modeled by digital computers, or by parallel-processing systems more like brains? Do computer programs consist of meaningless patterns, or do they embody (and explain) genuine meaning? (shrink)

In recent years, AI research has become more and more computationally demanding. In natural language processing, this tendency is reflected in the emergence of large language models like GPT-3. These powerful neural network-based models can be used for a range of NLP tasks and their language generation capacities have become so sophisticated that it can be very difficult to distinguish their outputs from human language. LLMs have raised concerns over their demonstrable biases, heavy environmental (...) footprints, and future social ramifications. In December 2020, critical research on LLMs led Google to fire Timnit Gebru, co-lead of the company’s AI Ethics team, which sparked a major public controversy around LLMs and the growing corporate influence over AI research. This article explores the role LLMs play in the political economy of AI as infrastructural components for AI research and development. Retracing the technical developments that have led to the emergence of LLMs, we point out how they are intertwined with the business model of big tech companies and further shift power relations in their favour. This becomes visible through the Transformer, which is the underlying architecture of most LLMs today and started the race for ever bigger models when it was introduced by Google in 2017. Using the example of GPT-3, we shed light on recent corporate efforts to commodify LLMs through paid API access and exclusive licensing, raising questions around monopolization and dependency in a field that is increasingly divided by access to large-scale computing power. (shrink)

As AI technologies are rolled out into healthcare, academia, human resources, law, and a multitude of other domains, they become de-facto arbiters of truth. But truth is highly contested, with many different definitions and approaches. This article discusses the struggle for truth in AI systems and the general responses to date. It then investigates the production of truth in InstructGPT, a large language model, highlighting how data harvesting, model architectures, and social feedback mechanisms weave together disparate understandings of veracity. (...) It conceptualizes this performance as an operationalization of truth, where distinct, often-conflicting claims are smoothly synthesized and confidently presented into truth-statements. We argue that these same logics and inconsistencies play out in Instruct’s successor, ChatGPT, reiterating truth as a non-trivial problem. We suggest that enriching sociality and thickening “reality” are two promising vectors for enhancing the truth-evaluating capacities of future language models. We conclude, however, by stepping back to consider AI truth-telling as a social practice: what kind of “truth” do we as listeners desire? (shrink)

The advent of ChatGPT by OpenAI has prompted extensive discourse on its potential implications for science and higher education. While the impact on education has been a primary focus, there is limited empirical research on the effects of large language models (LLMs) and LLM-based chatbots on science and scientific practice. To investigate this further, we conducted a Delphi study involving 72 researchers specializing in AI and digitization. The study focused on applications and limitations of LLMs, their effects on (...) the science system, ethical and legal considerations, and the required competencies for their effective use. Our findings highlight the transformative potential of LLMs in science, particularly in administrative, creative, and analytical tasks. However, risks related to bias, misinformation, and quality assurance need to be addressed through proactive regulation and science education. This research contributes to informed discussions on the impact of generative AI in science and helps identify areas for future action. (shrink)

Second-language learners rarely arrive at native proficiency in a number of linguistic domains, including morphological and syntactic processing. Previous approaches to understanding the different outcomes of first- versus second-language learning have focused on cognitive and neural factors. In contrast, we explore the possibility that children and adults may rely on different linguistic units throughout the course of language learning, with specific focus on the granularity of those units. Following recent psycholinguistic evidence for the role of multiword chunks (...) in online language processing, we explore the hypothesis that children rely more heavily on multiword units in language learning than do adults learning a second language. To this end, we take an initial step toward using large-scale, corpus-based computational modeling as a tool for exploring the granularity of speakers' linguistic units. Employing a computational model of language learning, the Chunk-Based Learner, we compare the usefulness of chunk-based knowledge in accounting for the speech of second-language learners versus children and adults speaking their first language. Our findings suggest that while multiword units are likely to play a role in second-language learning, adults may learn less useful chunks, rely on them to a lesser extent, and arrive at them through different means than children learning a first language. (shrink)

This book discusses how scientific and other types of cognition make use of models, abduction, and explanatory reasoning in order to produce important and innovative changes in theories and concepts. Gathering revised contributions presented at the international conference on Model-Based Reasoning, held on October 24–26 2018 in Seville, Spain, the book is divided into three main parts. The first focuses on models, reasoning, and representation. It highlights key theoretical concepts from an applied perspective, and addresses issues concerning information (...) visualization, experimental methods, and design. The second part goes a step further, examining abduction, problem solving, and reasoning. The respective papers assess different types of reasoning, and discuss various concepts of inference and creativity and their relationship with experimental data. In turn, the third part reports on a number of epistemological and technological issues. By analyzing possible contradictions in modern research and describing representative case studies, this part is intended to foster new discussions and stimulate new ideas. All in all, the book provides researchers and graduate students in the fields of applied philosophy, epistemology, cognitive science, and artificial intelligence alike with an authoritative snapshot of the latest theories and applications of model-based reasoning. (shrink)

Access to legal information is fundamental to access to justice. Yet accessibility refers not only to making legal documents available to the public, but also rendering legal information comprehensible to them. A vexing problem in bringing legal information to the public is how to turn formal legal documents such as legislation and judgments, which are often highly technical, to easily navigable and comprehensible knowledge to those without legal education. In this study, we formulate a three-step approach for bringing legal knowledge (...) to laypersons, tackling the issues of navigability and comprehensibility. First, we translate selected sections of the law into snippets (called CLIC-pages), each being a small piece of article that focuses on explaining certain technical legal concept in layperson’s terms. Second, we construct a Legal Question Bank, which is a collection of legal questions whose answers can be found in the CLIC-pages. Third, we design an interactive CLIC Recommender. Given a user’s verbal description of a legal situation that requires a legal solution, CRec interprets the user’s input and shortlists questions from the question bank that are most likely relevant to the given legal situation and recommends their corresponding CLIC pages where relevant legal knowledge can be found. In this paper we focus on the technical aspects of creating an LQB. We show how large-scale pre-trained language models, such as GPT-3, can be used to generate legal questions. We compare machine-generated questions against human-composed questions and find that MGQs are more scalable, cost-effective, and more diversified, while HCQs are more precise. We also show a prototype of CRec and illustrate through an example how our 3-step approach effectively brings relevant legal knowledge to the public. (shrink)

This paper delves into the striking parallels between the linguistic patterns of Large Language Models (LLMs) and the concepts of psychosis in Lacanian psychoanalysis. Lacanian theory, with its focus on the formal and logical underpinnings of psychosis, provides a compelling lens to juxtapose human cognition and AI mechanisms. LLMs, such as GPT-4, appear to replicate the intricate metaphorical and metonymical frameworks inherent in human language. Although grounded in mathematical logic and probabilistic analysis, the outputs of LLMs echo (...) the nuanced linguistic associations found in metaphor and metonymy, suggesting a mirroring of human linguistic structures. A pivotal point in this discourse is the exploration of “absolute metaphors”—core gaps in reasoning discernible in both AI models and human thought processes and central to the Lacanian conceptualization of psychosis. Despite the traditional divide between AI research and continental philosophy, this analysis embarks on an innovative journey, utilizing Lacanian philosophy to unravel the logic of AI, using concepts established in the continental discourse on logic, rather than the analytical tradition. (shrink)

In this paper, I argue for a modified version of what Devitt calls the Representational Thesis. According to RT, syntactic rules or principles are psychologically real, in the sense that they are represented in the mind/brain of every linguistically competent speaker/hearer. I present a range of behavioral and neurophysiological evidence for the claim that the human sentence processing mechanism constructs mental representations of the syntactic properties of linguistic stimuli. I then survey a range of psychologically plausible computational models (...) of comprehension and show that they are all committed to RT. I go on to sketch a framework for thinking about the nature of the representations involved in sentence processing. My claim is that these are best characterized not as propositional attitudes but, rather, as subpersonal states whose representational properties are determined by their functional role. Finally, I distinguish between explicit and implicit representations and argue that the latter can be drawn on as data by the algorithms that constitute our sentence processing routines. I conclude that skepticism concerning the psychological reality of grammars cannot be sustained. (shrink)

Transformer-based language models (TLMs) have widely been recognized to be a cutting-edge technology for the successful development of deep-learning-based solutions to problems and applications that require natural language processing and understanding. Like for other textual domains, TLMs have indeed pushed the state-of-the-art of AI approaches for many tasks of interest in the legal domain. Despite the first Transformer model being proposed about six years ago, there has been a rapid progress of this technology at an unprecedented rate, (...) whereby BERT and related models represent a major reference, also in the legal domain. This article provides the first systematic overview of TLM-based methods for AI-driven problems and tasks in the legal sphere. A major goal is to highlight research advances in this field so as to understand, on the one hand, how the Transformers have contributed to the success of AI in supporting legal processes, and on the other hand, what are the current limitations and opportunities for further research development. (shrink)

Mentalization describes the process through which we understand the mental states of oneself and others. In this paper, I present a computational semiotic model of mentalization and illustrate it through a worked-out example. The model draws on classical semiotic ideas, such as abductive inference and hypostatic abstraction, but pours them into new ideas and tools from natural language processing, machine learning, and neural networks, to form a novel model of language-mediated-mentalization.

Computational learning theory explores the limits of learnability. Studying language acquisition from this perspective involves identifying classes of languages that are learnable from the available data, within the limits of time and computational resources available to the learner. Different models of learning can yield radically different learnability results, where these depend on the assumptions of the model about the nature of the learning process, and the data, time, and resources that learners have access to. To the (...) extent that such assumptions accurately reflect human language learning, a model that invokes them can offer important insights into the formal properties of natural languages, and the way in which their representations might be efficiently acquired. In this chapter we consider several computational learning models that have been applied to the language learning task. Some of these have yielded results that suggest that the class of natural languages cannot be efficiently learned from the primary linguistic data (PLD) available to children, through.. (shrink)

The problem of computational complexity of semantics for some natural language constructions – considered in [M. Mostowski, D. Wojtyniak 2004] – motivates an interest in complexity of Ramsey quantifiers in finite models. In general a sentence with a Ramsey quantifier R of the following form Rx, yH(x, y) is interpreted as ∃A(A is big relatively to the universe ∧A2 ⊆ H). In the paper cited the problem of the complexity of the Hintikka sentence is reduced to the (...) problem of computational complexity of the Ramsey quantifier for which the phrase “A is big relatively to the universe” is interpreted as containing at least one representative of each equivalence class, for some given equvalence relation. In this work we consider quantifiers Rf, for which “A is big relatively to the universe” means “card(A) > f (n), where n is the size of the universe”. Following [Blass, Gurevich 1986] we call R mighty if Rx, yH(x, y) defines N P – complete class of finite models. Similarly we say that Rf is N P –hard if the corresponding class is N P –hard. We prove the following theorems. (shrink)

We compare our model of unsupervised learning of linguistic structures, ADIOS [1, 2, 3], to some recent work in computational linguistics and in grammar theory. Our approach resembles the Construction Grammar in its general philosophy (e.g., in its reliance on structural generalizations rather than on syntax projected by the lexicon, as in the current generative theories), and the Tree Adjoining Grammar in its computational characteristics (e.g., in its apparent affinity with Mildly Context Sensitive Languages). The representations learned by (...) our algorithm are truly emergent from the (unannotated) corpus data, whereas those found in published works on cognitive and construction grammars and on TAGs are hand-tailored. Thus, our results complement and extend both the computational and the more linguistically oriented research into language acquisition. We conclude by suggesting how empirical and formal study of language can be best integrated. (shrink)

Words are the essence of communication: They are the building blocks of any language. Learning the meaning of words is thus one of the most important aspects of language acquisition: Children must first learn words before they can combine them into complex utterances. Many theories have been developed to explain the impressive efficiency of young children in acquiring the vocabulary of their language, as well as the developmental patterns observed in the course of lexical acquisition. A major (...) source of disagreement among the different theories is whether children are equipped with special mechanisms and biases for word learning, or their general cognitive abilities are adequate for the task. We present a novel computational model of early word learning to shed light on the mechanisms that might be at work in this process. The model learns word meanings as probabilistic associations between words and semantic elements, using an incremental and probabilistic learning mechanism, and drawing only on general cognitive abilities. The results presented here demonstrate that much about word meanings can be learned from naturally occurring child-directed utterances (paired with meaning representations), without using any special biases or constraints, and without any explicit developmental changes in the underlying learning mechanism. Furthermore, our model provides explanations for the occasionally contradictory child experimental data, and offers predictions for the behavior of young word learners in novel situations. (shrink)

Molecular models are typical topics of chemical research depending on the technical standards of observation, computation, and representation. Mathematically, molecular structures have been represented by means of graph theory, topology, differential equations, and numerical procedures. With the increasing capabilities of computer networks, computational models and computer-assisted visualization become an essential part of chemical research. Object-oriented programming languages create a virtual reality of chemical structures opening new avenues of exploration and collaboration in chemistry. From an epistemic point of (...) view, virtual reality is a new computer-assisted tool of human imagination and recognition. (shrink)

Computational models of child language development can help us understand the cognitive underpinnings of the language learning process, which occurs along several linguistic levels at once (e.g., prosodic and phonological). However, in light of the replication crisis, modelers face the challenge of selecting representative and consolidated infant data. Thus, it is desirable to have evaluation methodologies that could account for robust empirical reference data, across multiple infant capabilities. Moreover, there is a need for practices that can (...) compare developmental trajectories of infants to those of models as a function of language experience and development. The present study aims to take concrete steps to address these needs by introducing the concept of comparing models with large-scale cumulative empirical data from infants, as quantified by meta-analyses conducted across a large number of individual behavioral studies. We formalize the connection between measurable model and human behavior, and then present a conceptual framework for meta-analytic evaluation of computational models. We exemplify the meta-analytic model evaluation approach with two modeling experiments on infant-directed speech preference and native/non-native vowel discrimination. (shrink)

Extract from Hofstadter's revew in Bulletin of American Mathematical Society : http://www.ams.org/journals/bull/1980-02-02/S0273-0979-1980-14752-7/S0273-0979-1980-14752-7.pdf -/- "Aaron Sloman is a man who is convinced that most philosophers and many other students of mind are in dire need of being convinced that there has been a revolution in that field happening right under their noses, and that they had better quickly inform themselves. The revolution is called "Artificial Intelligence" (Al)-and Sloman attempts to impart to others the "enlighten- ment" which he clearly regrets not having (...) experienced earlier himself. Being somewhat of a convert, Sloman is a zealous campaigner for his point of view. Now a Reader in Cognitive Science at Sussex, he began his academic career in more orthodox philosophy and, by exposure to linguistics and AI, came to feel that all approaches to mind which ignore AI are missing the boat. I agree with him, and I am glad that he has written this provocative book. The tone of Sloman's book can be gotten across by this quotation (p. 5): "I am prepared to go so far as to say that within a few years, if there remain any philosophers who are not familiar with some of the main developments in artificial intelligence, it will be fair to accuse them of professional incom- petence, and that to teach courses in philosophy of mind, epistemology, aesthetics, philosophy of science, philosophy of language, ethics, metaphysics, and other main areas of philosophy, without discussing the relevant aspects of artificial intelligence will be as irresponsible as giving a degree course in physics which includes no quantum theory." -/- (The author now regrets the extreme polemical tone of the book.). (shrink)