Rebekah A. Gelpí, Yibing Ju, Ethan C. Jackson
et al.
We introduce Sorrel (https://github.com/social-ai-uoft/sorrel), a simple Python interface for generating and testing new multi-agent reinforcement learning environments. This interface places a high degree of emphasis on simplicity and accessibility, and uses a more psychologically intuitive structure for the basic agent-environment loop, making it a useful tool for social scientists to investigate how learning and social interaction leads to the development and change of group dynamics. In this short paper, we outline the basic design philosophy and features of Sorrel.
While memes are often humorous, they are frequently used to disseminate hate, causing serious harm to individuals and society. Current approaches to hateful meme detection mainly rely on pre-trained language models. However, less focus has been dedicated to \textit{what make a meme hateful}. Drawing on insights from philosophy and psychology, we argue that hateful memes are characterized by two essential features: a \textbf{presupposed context} and the expression of \textbf{false claims}. To capture presupposed context, we develop \textbf{PCM} for modeling contextual information across modalities. To detect false claims, we introduce the \textbf{FACT} module, which integrates external knowledge and harnesses cross-modal reference graphs. By combining PCM and FACT, we introduce \textbf{\textsf{SHIELD}}, a hateful meme detection framework designed to capture the fundamental nature of hate. Extensive experiments show that SHIELD outperforms state-of-the-art methods across datasets and metrics, while demonstrating versatility on other tasks, such as fake news detection.
The relevance of the study is due to the necessity to analyze mutual determination of the development of language and culture so as to expand the problematic field of humanities. Thus, the article is focused on Alexei Losev’s book Philosophy of Name (1927) characterized with a kind of polyphony that calls for bridging the author’s position with a variety of philosophical trends dealing with the ontological problems of language and culture. The article’s authors set the goal of examining the specifics of studies on ontological characteristics of language through the prism of its cultural representation, analysing Losev’s conceptual framework. In this regard the authors set the following tasks: firstly, to determine the pre-object structure of the name as a mythologeme, by highlighting the primary status of the advancement from the internal sense, related to the existential foundations of life, to its external representations; secondly, to analyze the reverse motion from essence to its appearances as realized in the word expressing the fullness of individual and cultural meanings. In accordance with the set tasks, the results of the study follow the general logic of Losev’s conceptualizations. The dialectic of Losev’s reasoning combines the appraisal of the ascent to the Absolute in understanding the name, on the one hand, with the significance of the descent to the word with its agile lively content, on the other. Losev integrates these two trends in language development through myth, which he sees as a living reality—an actual manifestation of essence fixed in a name. The conclusions are indicated by appealing to the ontological Pan-Unity allowing Losev to integrate the ideas of man and the world, to connect reality to thought in the context of universal relations the wholeness of which becomes real in the living space of culture.
Do large language models (LLMs) exhibit any forms of awareness similar to humans? In this paper, we introduce AwareBench, a benchmark designed to evaluate awareness in LLMs. Drawing from theories in psychology and philosophy, we define awareness in LLMs as the ability to understand themselves as AI models and to exhibit social intelligence. Subsequently, we categorize awareness in LLMs into five dimensions, including capability, mission, emotion, culture, and perspective. Based on this taxonomy, we create a dataset called AwareEval, which contains binary, multiple-choice, and open-ended questions to assess LLMs' understandings of specific awareness dimensions. Our experiments, conducted on 13 LLMs, reveal that the majority of them struggle to fully recognize their capabilities and missions while demonstrating decent social intelligence. We conclude by connecting awareness of LLMs with AI alignment and safety, emphasizing its significance to the trustworthy and ethical development of LLMs. Our dataset and code are available at https://github.com/HowieHwong/Awareness-in-LLM.
Christian de Ronde, Raimundo Fernández Mouján, César Massri
In this work we argue against the interpretation that underlies the "Standard" account of Quantum Mechanics (SQM) that was established during the 1930s by Niels Bohr and Paul Dirac. Ever since, following this orthodox narrative, physicists have dogmatically proclaimed -- quite regardless of the deep contradictions and problems -- that the the theory of quanta describes a microscopic realm composed of elementary particles (such as electrons, protons and neutrons) which underly our macroscopic world composed of tables, chairs and dogs. After critically addressing this atomist dogma still present today in contemporary (quantum) physics and philosophy, we present a new understanding of quantum individuals defined as the minimum set of relations within a specific degree of complexity capable to account for all relations within that same degree. In this case, quantum individuality is not conceived in absolute terms but -- instead -- as an objectively relative concept which even though depends of the choice of bases and factorizations remain nonetheless part of the same invariant representation.
We take supersymmetry in the Seiberg-Witten theory as a case study of the uses of (super)symmetry arguments in studying the ontology of four-dimensional interacting quantum field theories. Together with a double expansion, supersymmetry is a via media that helps to bridge the gap between the ontologies of an exact quantum field theory and its semi-classical limit. We discuss a class of states that exist at any value of the coupling, and whose properties such as mass, electric and magnetic charges, and spin quantum numbers can be precisely characterised at low energies. The low-energy theory is best presented as a one-dimensional complex manifold, equipped with metric and other structures: namely, the space of low-energy vacua, covered by three open regions that are interpreted as macroscopic phases. We discuss two cases of emergence: the emergence of the low-energy regime and the emergence between models at low energies, thereby highlighting the significance of the topology of the space of vacua for such cases of emergence.
Humans and machines interact more frequently than ever and our societies are becoming increasingly hybrid. A consequence of this hybridisation is the degradation of societal trust due to the prevalence of AI-enabled deception. Yet, despite our understanding of the role of trust in AI in the recent years, we still do not have a computational theory to be able to fully understand and explain the role deception plays in this context. This is a problem because while our ability to explain deception in hybrid societies is delayed, the design of AI agents may keep advancing towards fully autonomous deceptive machines, which would pose new challenges to dealing with deception. In this paper we build a timely and meaningful interdisciplinary perspective on deceptive AI and reinforce a 20 year old socio-cognitive perspective on trust and deception, by proposing the development of DAMAS -- a holistic Multi-Agent Systems (MAS) framework for the socio-cognitive modelling and analysis of deception. In a nutshell this paper covers the topic of modelling and explaining deception using AI approaches from the perspectives of Computer Science, Philosophy, Psychology, Ethics, and Intelligence Analysis.
The paper discusses the possible implications of the relational framework of Pure Shape Dynamics for the metaphysics of time. The starting point of the analysis is an interpretation of shapes in ontic structural realist terms, which gives rise to the notion of self-subsisting structure. The relational version of a Newtonian-particle toy model is introduced and discussed as a concrete example.
Bruno Latour, ‘I am interested in Europe as an ecological problem’ (2019): - Tony Judt, ‘The Way Things Are and How They Might Be’, <em> London Review of Books </em> (2010) <em> : </em> - Sam Knight, ‘What Have Fourteen Years of Conservative Rule Done to Britain?’ <em> The New Yorker </em> (2024): - Salman Rushdie, <em> Knife: Meditations After an Attempted Murder </em> (2024): - Adam Curtis, Interview with <em> Jacobin </em> (2023):
We propose an interpretation of physics named potentiality realism. This view, which can be applied to classical as well as to quantum physics, regards potentialities (i.e. intrinsic, objective propensities for individual events to obtain) as elements of reality, thereby complementing the actual properties taken by physical variables. This allows one to naturally reconcile realism and fundamental indeterminism in any theoretical framework. We discuss our specific interpretation of propensities, that require them to depart from being probabilities at the formal level, though allowing for statistics and the law of large numbers. This view helps reconcile classical and quantum physics by showing that most of the conceptual problems that are customarily taken to be unique issues of the latter -- such as the measurement problem -- are actually in common to all indeterministic physical theories.
Friedrich Nietzsche, <em> Beyond Good and Evil </em> : - Leo Tolstoy, <em> War and Peace </em> : - Christopher Clark, <em> The Sleepwalkers: How Europe Went to War in 1914 </em> : - Lawrence Freedman, <em> Strategy: A History </em> : - Stefan Zweig, <em> The World of Yesterday </em> : - Albert Camus, <em> The Rebel </em> : - Karl Jaspers, <em> Tragedy is Not Enough </em> : - Innokenty Annensky, ‘Nightmares’:
In Logik der Forschung and later works, Karl Popper proposed a set of methodological rules for scientists. Among these were requirements that theories evolve in the direction of increasing content, and that new theories should only be accepted if some of their novel predictions are experimentally confirmed. There are currently two, viable theories of cosmology: the standard cosmological model, and a theory due to Mordehai Milgrom called MOND. Both theories can point to successes and failures, but only MOND has repeatedly made novel predictions that were subsequently found to be correct. Standard-model cosmologists, by contrast, have almost always responded to new observations in a post-hoc manner, adjusting or augmenting their theory as needed to obtain correspondence with the facts. I argue that these methodological differences render a comparison of the two theories in terms of their "truthlikeness" or "verisimilitude" essentially impossible since the two groups of scientists achieve correspondence with the facts in often very different ways, and I suggest that a better guide to the theories' progress toward truth might be the methodologies themselves.
Imagining the real. Really imagining. There and here, blending and blurring, all together. Hayao Miyazaki, <em> Shuna’s Journey. </em> Christopher de Bellaigue on the possibilities of ‘an unstoppable spiral of state violence and popular fury’ in Iran. Alexander Baunov on Russia’s objectives. 2022 is set to be ‘a fabulous year’ for some. Ali Ansari on ‘failures of imagination’ in Iran.
This paper is an interview with Vladimir M. Alpatov, Doctor of Philology, full member of the Russian Academy of Sciences, distinguished linguist, and expert on Japanese studies. The focus of the interview is intercultural communication and the problems associated with translating languages and cultural realities. The difference between cultures makes it necessary to provide extensive commentaries to make the text comprehensive to foreign readers. Though, Vladimir M. Alpatov notes, the comments depend on the purposes and types of translation, such as academic or literary translation. Symbols are part of a greater symbolic and cultural system. Often it is not the symbol but one’s attitude towards the object that causes misunderstanding and requires clarification. Vladimir A. Alpatov gives many examples of how the Japanese view and treat life differently from Russian people. Many discrepancies come from domestic life and economic practices: the Japanese are less knowledgeable about cattle than many other nations. At the same time, insects that are found all around the world receive special treatment and admiration. Vladimir A. Alpatov makes a critical point on the absence of a proper method of studying cultural differences. We observe and list numerous cultural differences, but explanations and theories we come up with have no solid methodological basis. Another topic discussed is machine translation and AI Linguistics used to be considered exact science that implied the possibility of machine translation not assisted by humans. However, it did not happen yet, and the need for human-to-human translation or post-editing is obvious. With literary translation and translation from unrelated languages, the case against AI is stronger — human intuition in translating cultural specifics is indispensable, and various translations rather than a single canonic one should be welcome. Differentiation of sciences brought about cultural studies and linguistic-cultural studies that finally embraced the study of language as one of the vital elements of culture. Today many students study foreign languages and are interested in intercultural communication. They need to learn that we can overcome bias and prejudices through personal contact. One more way to promote a different vision of one’s culture and country — is to speak about it in an understandable language, for example, on the Internet.
Bryan W. Roberts, Henrique Gomes, Jeremy Butterfield
Why is gauge symmetry so important in modern physics, given that one must eliminate it when interpreting what the theory represents? In this paper we discuss the sense in which gauge symmetry can be fruitfully applied to constrain the space of possible dynamical models in such a way that forces and charges are appropriately coupled. We review the most well-known application of this kind, known as the 'gauge argument' or 'gauge principle', discuss its difficulties, and then reconstruct the gauge argument as a valid theorem in quantum theory. We then present what we take to be a better and more general gauge argument, based on Noether's second theorem in classical Lagrangian field theory, and argue that this provides a more appropriate framework for understanding how gauge symmetry helps to constrain the dynamics of physical theories.
In his book `Physics and Philosophy', Heisenberg suggested that the quantum world is one of ``potentialities or possibilities'' and that the classical realm is one of ``things or facts''. After ascertaining that his categories most naturally have the structure of ontological equivalence classes, we show that they cannot be seriously introduced into the quantum formalism without rendering it incoherent, as under some circumstances the formalism permits an equality between members of distinct equivalence classes. This is labeled the `equality of inequivalents problem'. Three possible reactions are discussed: First, one could deny the problem by challenging some of the assumptions that underlie its formulation; second, one could accept the problem and take it to be grounds for dismissing Heisenberg's or similar distinctions altogether; third, one could accept the problem but take its source to be not Heisenberg's distinction but the quantum formalism. A plausibility argument in support of the third reaction is given by analogy: it is shown that axiomatic probability suffers from a similar problem if one wishes to seriously distinguish between things which are possibilities and things which are not. An enriched axiomatization of probability is proposed which captures the concept of probability as a measure over possibilities and thereby overcomes the equality of inequivalents problem in axiomatic probability.
The article observes public speaking skills as a socio-cultural phenomenon in the traditional Abkhaz society at national and individual levels. The topicality of the study lies in analyzing certain functions and genre modifications of oratory tradition among Abkhazians. In our considered opinion, public speaking is an integral element of the life of Abkhaz people, an essential phenomenon for understanding the Abkhazian worldview, their culture and attitude to reality. Such analysis might be vital for the reconstruction of a holistic «portrait» of the Abkhazian ethnicity, understanding the character, intentions, mentality of its representatives. The aim of the study is to analyze the specific functions of oratory tradition for the Abkhaz society. The task is to classify its main functions; to consider genre modifications of oratory tradition aimed at harmonization of the society; to determine the role and importance of the speaker in the traditional Abkhaz society. On the basis of the research conducted, the author comes to the conclusion that in a traditional Abkhaz society is oratory performed a number of socially significant regulatory and ideological functions, namely, ordered intercultural interactions, coordinated the resolution of political and socially important issues, worked as a value-regulating mechanism. At an individual level, by mastering the art of speaking in public, Abkhazians realized natural social needs, the need for communication, socialization and creative activities.
This work presents parallel histories of the development of two modern theories of condensed matter: the theory of electron structure in quantum mechanics, and the theory of liquid structure in statistical mechanics. Comparison shows that key revelations in both are not only remarkably similar, but even follow along a common thread of controversy that marks progress from antiquity through to the present. This theme appears as a creative tension between two competing philosophies, that of short range structure (atomistic models) on the one hand, and long range structure (continuum or density functional models) on the other. The timeline and technical content are designed to build up a set of key relations as guideposts for using density functional theories together with atomistic simulation.
We show how faceted search using a combination of traditional classification systems and mixed-membership topic models can go beyond keyword search to inform resource discovery, hypothesis formulation, and argument extraction for interdisciplinary research. Our test domain is the history and philosophy of scientific work on animal mind and cognition. The methods can be generalized to other research areas and ultimately support a system for semi-automatic identification of argument structures. We provide a case study for the application of the methods to the problem of identifying and extracting arguments about anthropomorphism during a critical period in the development of comparative psychology. We show how a combination of classification systems and mixed-membership models trained over large digital libraries can inform resource discovery in this domain. Through a novel approach of "drill-down" topic modeling---simultaneously reducing both the size of the corpus and the unit of analysis---we are able to reduce a large collection of fulltext volumes to a much smaller set of pages within six focal volumes containing arguments of interest to historians and philosophers of comparative psychology. The volumes identified in this way did not appear among the first ten results of the keyword search in the HathiTrust digital library and the pages bear the kind of "close reading" needed to generate original interpretations that is the heart of scholarly work in the humanities. Zooming back out, we provide a way to place the books onto a map of science originally constructed from very different data and for different purposes. The multilevel approach advances understanding of the intellectual and societal contexts in which writings are interpreted.
Recent tremendous development of quantum information theory led to a number of quantum technological projects, e.g., quantum random generators. This development stimulates a new wave of interest in quantum foundations. One of the most intriguing problems of quantum foundations is elaboration of a consistent and commonly accepted interpretation of quantum state. Closely related problem is clarification of the notion of quantum randomness and its interrelation with classical randomness. In this short review we shall discuss basics of classical theory of randomness (which by itself is very complex and characterized by diversity of approaches) and compare it with irreducible quantum randomness. The second part of this review is devoted to the information interpretation of quantum mechanics (QM) in the spirit of Zeilinger and Brukner (and QBism of Fuchs et al.) and physics in general (e.g., Wheeler's "it from bit") as well as digital philosophy of Chaitin (with historical coupling to ideas of Leibnitz). Finally, we continue discussion on interrelation of quantum and classical randomness and information interpretation of QM.