The transformers that drive chatbots and other AI systems constitute large language models (LLMs). These are currently the focus of a lively discussion in both the scientific literature and the popular media. This discussion ranges from hyperbolic claims that attribute general intelligence and sentience to LLMs, to the skeptical view that these devices are no more than “stochastic parrots”. I present an overview of some of the weak arguments that have been presented against LLMs, and I consider several of the (...) more compelling criticisms of these devices. The former significantly underestimate the capacity of transformers to achieve subtle inductive inferences required for high levels of performance on complex, cognitively significant tasks. In some instances, these arguments misconstrue the nature of deep learning. The latter criticisms identify significant limitations in the way in which transformers learn and represent patterns in data. They also point out important differences between the procedures through which deep neural networks and humans acquire knowledge of natural language. It is necessary to look carefully at both sets of arguments in order to achieve a balanced assessment of the potential and the limitations of LLMs. (shrink)

Was Maimonides a radical philosopher who subtly argued for a naturalist world and who saw the obligation to keep the Torah's commandments as a social and moral obligation - or was he a conservative Jewish believer who only tried to formulate philosophical arguments in favour of a revealed religion? This question has been central to the interpretation of Maimonides from the 12th century until modern times. In the four chapters of this book, Shalom Sadik argues for a radical philosophical (...) interpretation of Maimonides. (shrink)

1. Formal semantics in linguistics -- 2. Generalized quantifier theory -- 3. The interface between syntax and semantics -- 4. Anaphora, discourse, and modality -- 5. Focus, presupposition, and negation -- 6. Tense -- 7. Questions -- 8. Plurals -- 9. Computational semantics -- 10. Lexical semantics -- 11. Semantics and related domains.

The application of deep learning methods to problems in natural language processing has generated significant progress across a wide range of natural language processing tasks. For some of these applications, deep learning models now approach or surpass human performance. While the success of this approach has transformed the engineering methods of machine learning in artificial intelligence, the significance of these achievements for the modelling of human learning and representation remains unclear. Deep Learning and Linguistic Representation looks at the application of (...) a variety of deep learning systems to several cognitively interesting NLP tasks. It also considers the extent to which this work illuminates our understanding of the way in which humans acquire and represent linguistic knowledge. Key Features: combines an introduction to deep learning in AI and NLP with current research on Deep Neural Networks in computational linguistics. is self-contained and suitable for teaching in computer science, AI, and cognitive science courses; it does not assume extensive technical training in these areas. provides a compact guide to work on state of the art systems that are producing a revolution across a range of difficult natural language tasks. (shrink)

Toldot adam 3 -- Bet ha-midot -- Sheʼelot u-teshuvot.

In this paper, we explore the possibility that machine learning approaches to naturallanguage processing being developed in engineering-oriented computational linguistics may be able to provide specific scientific insights into the nature of human language. We argue that, in principle, machine learning results could inform basic debates about language, in one area at least, and that in practice, existing results may offer initial tentative support for this prospect. Further, results from computational learning theory can inform arguments carried on within linguistic theory (...) as well. (shrink)

Sortally incorrect sentences have traditionally been referred to as "category mistakes" (Ryle ()) or "type crossings" (Drange ()). Sortal incorrectness is a ...

Raquel FERN ´ ANDEZ, Jonathan GINZBURG and Shalom LAPPIN Department of Computer Science King’s College London Strand, London WC2R 2LS, UK {raquel,ginzburg,lappin}@dcs.kcl.ac.uk..

In lieu of an abstract, here is a brief excerpt of the content:Reply to Russell's letter of 16 May 1960 by Albert Shalom EDITORIAL NOTE To illustrate a list ofrecent acquisitions in Russell (Summer 1981), we printed in facsimile Russell's letter of 16 May 1960 to Professor Albert Shalom concerning the interpretation of Wittgenstein's Tractatus LogicoPhilosophicus. The correspondence between Russell and Shalom began when Shalom wrote on I May 1960 asking whether Russell had the time and (...) inclination to read a translation of a lecture which Shalom had given the previous December in Paris. The paper was entitled "The Metaphysical Thinking Underlying Wittgenstein's Tractatus". Shalom had received "different and even contradictory appraisals" of his paper, and he very much wanted to have Russell's opinion of it. Close to the age ofninety, Russell was entering a hectic period in his life in which he worked assiduouslyfor the Campaign for Nuclear Disarmament. He had published three books in 1959 and had produced prefaces and introductions for other people. As usual, he was besieged with mail including numerous requests to read unsolicited manuscripts sent to him quite regularly. Yet he had managed to answer correspondence in a very personal way. In November 1959, he had been involved in a controversy regarding the refusal ofthe editor of Mind to review Ernest Gellner's Words and Things, a book which attacked the then prevalent philosophy of linguistic analysis and for which Russell had written the introduction. In the early months of1960, he took time out from his usual writing on the nuclear threat to reply to several articles devoted to his theory of descriptions and his ethical theory in the journal Philosophy. Russell's reply (4 May 1960) to Shalom's first letter states: "Thank you for your letter ofMay I. I shall be interested to see your paper on Wittgenstein's Tractatus, but, as I am very busy, it may be some time before I can read it carefully." On 8 May Shalom wrote back in appreciation enclosing his paper and an offprint of an article, "Qu'est-ce qu'un concept?", Revue internationale dephilosophie, 50 (1959): I-IS. The main objection raised against his paper, Shalom admitted, was that Wittgenstein was not concerned with problems of epistemology at the time of his writing the Tractatus. Nonetheless, Shalom felt that the "picture theory" of reality implies an epistemology, and he was unable to see a fallacy in his paper: "Perhaps the fact that I'm not a logician has something to do with the matter. Ifyou could enlighten me on this point I should be most grateful to you." 45 28th April 1982. 46 Russell winter 1982-83 Russell's letter of 16 May followed. On 25 May Shalom challenged Russell's interpretation. Far /rom underestimating the notion of structure, Shalom claimed that he had emphasized the notion as the key to understanding the Tractatus. "In fact, my main point is that Wittgenstein's analyses of various concepts are unintelligible unless understood as arguments in terms of the dominating idea ofstructure or form. And that is what I understand as a metaphysical argument." Shalom went on to say that his interest in any philosopher was to try to discover the chief concepts employed in the philosopher's analysis and to examine the manner in which the analysis was performed. For this reason, the later Wittgenstein was not devoid ofinterest. In a postscript, he asked Russell whether he would be in London in July ofthat year, and ifso, whether a meeting was possible. Russell replied on 18June as follows: "Thank you for your letter ofMay 25. I am sorry that at present my time and thought are so occupied with matters very remote from those with whichyour letter deals that I cannot offeryou a reasoned reply. Unfortunately, I do not expect to be in London in July so that I am afraid a meeting will hardly be possible." On 14 Shalom requested the return of his paper, and Russell complied on 19 July. Shalom is now Professor ofPhilosophy at McMaster. We invited him to comment on his exchange with Russell, particularly Russell's letter of16 May 1960. His comments take the form ofa further (and no doubtfinal) letter in the exchange.-Carl... (shrink)

Do innocent civilians who will be killed in a justified attack on a nearby military target have a right to defend themselves by shooting down the bomber pilot? I argue that they do not, and that Jeff McMahan's view that they do have such a right—that there is a moral equivalence between pilot and civilian—is flawed in much the same way that Michael Walzer's moral equivalence of combatants—a position that McMahan has so persuasively refuted—is flawed.

Machine learning and statistical methods have yielded impressive results in a wide variety of natural language processing tasks. These advances have generally been regarded as engineering achievements. In fact it is possible to argue that the success of machine learning methods is significant for our understanding of the cognitive basis of language acquisition and processing. Recent work in unsupervised grammar induction is particularly relevant to this issue. It suggests that knowledge of language can be achieved through general learning procedures, and (...) that a richly articulated language faculty is not required to explain its acquisition. (shrink)

Following participation in Dr. Yair Caspi’s “Psychology in Judaism” workshop, the writers contemplate whether the teaching of Caspi’s model in academic settings could become simultaneously a fresh addition to interdisciplinary approaches to the teaching of Judaism in Israeli Academic life, and an academic addition to the contemporary trend to Jewish renewal in Israeli society. The model is based on weekly facilitated workshops in which participants both reflect on and discuss their lives and also explore unique interpretations of Jewish texts and (...) ideas, constantly seeking the borderline where their lives inform the text and vice versa. The process leads to the re-enchantment of ancient Jewish concepts, the reinvention of the Jewish religious experience in a contemporary paradigm and idiom, and the possibility of deep transformative insight. (shrink)

The human self model suggests that the construct of self involves functions such as agency, body-centered spatial perspectivity, and long-term unity. Vogeley, Kurthen, Falkai, and Maieret (1999) suggest that agency is subserved by the prefrontal cortex and other association areas of the cortex, spatial perspectivity by the prefrontal cortex and the parietal lobes, and long-term unity by the prefrontal cortex and the temporal lobes and that all of these functions are impaired in schizophrenia. Exploring the connections between the prefrontal cortex (...) and the construct of self, the present article extends the application of the self model to autism. It suggests that in contrast to schizophrenia, agency and spatial perspectivity are probably preserved in autism, but that, similarly to schizophrenia, long-term unity is probably impaired. This hypothesis is compatible with a model of neuropsychological dysfunction in autism in a neural network including parts of the prefrontal cortex, the temporal lobes, and the cerebellum. (shrink)

We formulate a Curry-typed logic with fine-grained intensionality within Turner’s typed predicate logic. This allows for an elegant presentation of a theory that corresponds to Fox and Lappin’s property theory with curry typing, but without the need for a federation of languages. We then consider how the fine-grained intensionality of this theory can be given an operational interpretation. This interpretation suggests itself as expressions in the theory can be viewed as terms in the untyped lambda-calculus, which provides a model of (...) computation. (shrink)

On the Fregean view of NP's, quantified NP's are represented as operator-variable structures while proper names are constants appearing in argument position. The Generalized Quantifier approach characterizes quantified NP's and names as elements of a unified syntactic category and semantic type. According to the Logicality Thesis, the distinction between quantified NP's, which undergo an operation of quantifier raising to yield operator-variable structures at Logical Form and non-quantified NP's, which appear in situ at LF, corresponds to a difference in logicality status. (...) The former are logical expressions while the latter are not. Using van Benthem's [2, 3] criterion for logicality, I extend the concept of logicality to GQ's. I argue that NP's modified by exception phrases constitute a class of quantified NP's which are heterogeneous with respect to logicality. However, all exception phrase NP's exhibit the syntactic and semantic properties which motivate May to treat quantified NP's as operators at LF. I present a semantic analysis of exception phrases as modifiers of GQ's, and I indicate how this account captures the central semantic properties of exception phrase NP's. I explore the consequences of the logically heterogeneous character of exception phrase NP's for proof theoretic accounts of quantifiers in natural language. The proposed analysis of exception phrase NP's provides support for the GQ approach to the syntax and semantics of NP's. (shrink)

Sewards and Sewards argue that while computations necessary for object recognition occur throughout the ventral visual stream, object recognition awareness involves the anterior temporal lobe and the medial orbital prefrontal cortex. The present paper suggests, however, that the medial orbital prefrontal cortex has a unique contribution, namely that of producing a basic experience of a perceptual object. It is further argued that the mechanisms that produce this experience also result in making the object more important than its subparts and features. (...) Finally, it is argued that a reduction in this importance may account for some perceptual difficulties in high-functioning autism. This view is consistent with evidence for early selective abnormalities in other systems involving the medial prefrontal cortex in autism. (shrink)

Le but de ces quelques pages est de clarifier la manière dont Sartre utilise la notion de «l'être» dans son principal ouvrage, l'Être et le Néant. L'être, nous dit Sartre, est « dévoilé par quelque moyen d'accès immédiat» qu'il nomme « l'ennui», « la nausée » etc. « L'être » dont il est ici question est, nous le verrons, le concept général qui s'applique aux deux catégories d'existents qui, selon Sartre, composent le monde: l'être-en-soi et l'être-pour-soi. On a souvent dit (...) que l'intérêt de cet ouvrage se trouve dans les analyses qu'il nous donne de l'être-pour-soi qui est le nom donné à la « réalité humaine », et non dans ce qu'il nous dit concernant l'être-en-soi. II est vrai que Sartre ne consacre que quelques pages à cette derniére question. Cependant, il me semblc que ce que nous dit Sartre concernant l'être-en-soi est la clef de son ontologie de base, et que le sens de cet « En-soi» éclairera, également, certaines de ses affirmations concernant le « Pour-soi». Les deux concepts sont solidaires, mais le sens de l'En-soi est, en quelque sorte, plus fondamental que le sens du Pour-soi. (shrink)

Este artículo examina los diversos niveles de certeza en el pensamiento de Hasdai Crescas, tal como pueden encontrarse en su _Sermón de Pascua_ y su principal obra filosófica, _Or Hashem_. En la primera parte del artículo se analiza la discusión principal de Crescas sobre este tema en el _Sermón._ En la segunda parte, las principales ideas expuestas allí se comparan con la opinión de Crescas en _Or Hashem_. Además, el artículo busca explicar por qué este tema está más desarrollado en (...) el _Sermón_ que en su principal obra filosófico. En última instancia, se concluye que Hasdai Crescas identifica cuatro niveles distintos de certeza: (1) pruebas irrefutables y la experiencia de la profecía; (2) pruebas lógicas razonables y la presencia de milagros que no tienen explicación terrenal; (3) pruebas y milagros menos probables de naturaleza dudosa; y (4) aceptar el significado claro del texto de la Torá y la tradición en general. Es probable que R. Crescas no mencionara este tema en _Or Hashem_ debido al bajo grado de certeza de la tradición, que eligió mantener oculta. (shrink)

This book explores what motivates net-generation community college students to complete their degrees. By emphasizing relationships, personal growth, and support systems, the author’s research will empower educational institutions to increase retention and foster success.

The language of standard propositional modal logic has one operator (? or ?), that can be thought of as being determined by the quantifiers ? or ?, respectively: for example, a formula of the form ?F is true at a point s just in case all the immediate successors of s verify F.This paper uses a propositional modal language with one operator determined by a generalized quantifier to discuss a simple connection between standard invariance conditions on modal formulas and generalized (...) quantifiers: the combined generalized quantifier conditions of conservativity and extension correspond to the modal condition of invariance under generated submodels, and the modal condition of invariance under bisimulations corresponds to the generalized quantifier being a Boolean combination of ? and ? (shrink)

Studies in Abraham Joshua Heschel's Oeuvre.

The capacity to recognise and interpret sluices—bare wh-phrases that exhibit a sentential meaning—is essential to maintaining cohesive interaction between human users and a machine interlocutor in a dialogue system. In this paper we present a machine learning approach to sluice disambiguation in dialogue. Our experiments, based on solid theoretical considerations, show that applying machine learning techniques using a compact set of features that can be automatically identified from PoS markings in a corpus can be an efficient tool to disambiguate between (...) sluice interpretations. (shrink)

I compare several types of knowledge-based and knowledge-poor approaches to anaphora and ellipsis resolution. The former are able to capture fine-grained distinctions that depend on lexical meaning and real world knowledge, but they are generally not robust. The latter show considerable promise for yielding wide coverage systems. However, they consistently miss a small but significant subset of cases that are not accessible to rough-grained techniques of intepretation. I propose a sequenced model which first applies the most computationally efficient and inexpensive (...) methods to resolution and then progresses successively to more costly techniques to deal with cases not handled by previous modules. Confidence measures evaluate the judgements of each component in order to determine which instances of anaphora or ellipsis are to be passed on to the next, more fine-grained subsystem. (shrink)