Commonsense Concepts Research Articles

Learning Commonsense-aware Moment-Text Alignment for Fast Video Temporal Grounding

Grounding temporal video segments described in natural language queries effectively and efficiently is a crucial capability needed in vision-and-language fields. In this paper, we deal with the fast video temporal grounding (FVTG) task, aiming at localizing the target segment with high speed and favorable accuracy. Most existing approaches adopt elaborately designed cross-modal interaction modules to improve the grounding performance, which suffer from the test-time bottleneck. Although several common space-based methods enjoy the high-speed merit during inference, they can hardly capture the comprehensive and explicit relations between visual and textual modalities. In this paper, to tackle the dilemma of speed-accuracy tradeoff, we propose a commonsense-aware cross-modal alignment network (C 2 AN), which incorporates commonsense-guided visual and text representations into a complementary common space for fast video temporal grounding. Specifically, the commonsense concepts are explored and exploited by extracting the structural semantic information from a language corpus. Then, a commonsense-aware interaction module is designed to obtain bridged visual and text features by utilizing the learned commonsense concepts. Finally, to maintain the original semantic information of textual queries, a cross-modal complementary common space is optimized to obtain matching scores for performing FVTG. Extensive results on two challenging benchmarks show that our C 2 AN method performs favorably against state-of-the-arts while running at high speed. Our code is available at https://github.com/ZiyueWu59/CCA.

The Methodologically Flawed Discussion about Deep Disagreement

AbstractQuestions surrounding deep disagreement have gained significant attention in recent years. One of the central debates is metaphysical, focusing on the features that make a disagreement deep. Proposals for what makes disagreements deep include theories about hinge propositions and first epistemic principles. In this paper, I criticize this metaphysical discussion by arguing that it is methodologically flawed. Deep disagreement is a technical or semi-technical term, but the metaphysical discussion mistakenly treats it as a common-sense concept to be analyzed and captured by our pre-theoretical intuitions. Since the literature on deep disagreement is subject to this fundamental confusion and deep disagreement is not a helpful umbrella term either, I propose eliminating the notion of deep disagreement from the philosophical discourse. Instead of analyzing the nature of deep disagreement, we should develop theories about different forms of disagreement, including disagreement about hinge propositions and disagreement about epistemic principles, and, in particular, a theory of rationally irresolvable disagreement.

Kitchen Table Politics: Bannock and Métis Common Sense in an Era of Nascent Recognition Politics

Abstract: Bannock, a simple bread made of water, flour, and lard—fried or baked—is a staple of Indigenous diets across what is now called Canada. A pan-Indigenous symbol, bannock is a historically dynamic food grounded in both European and Indigenous origins. On both counts, it presents a paradox to the settler imagination, which clings to fixed definitions of Indigenous Peoplehood essentialized in precontact traditions. For Métis, however, bannock is no paradox. Neither its European origins nor its diverse forms and composition across time and place cause confusion. Rather, in oral history interviews Métis positioned bannock as a critical component that sustained a Métis identity through the twentieth century. Bannock offers important lessons for understanding the place of Métis within Canadian history and reveals how Métis mediated state interventions into Indigeneity in the 1980s. Tracing this historical trajectory, I suggest a useful inversion of Mark Rifkin's concept of settler common sense to focus on what I call a Métis common sense; that is, those aspects of a Métis livedness that were obvious for Métis. I follow other Métis writers who have proposed the kitchen table as a site of Métis identity survivance that functions as an alternative to public, androcentric expressions of Métis-ness legible to Canadian recognition politics. Métis interviewees negotiated with, and simultaneously rejected, essentialist assumptions of their Indigeneity. Interviewees understood bannock as a key marker of kinship sustained through female labor and activism within a matrilocal Métis Peoplehood.

Common sense and the difference between natural and human sciences

ABSTRACT This article proposes a new account of the relation between the sciences and common sense. A debate between Alfred North Whitehead and Arthur S. Eddington highlighted both the tendency of the natural sciences to repudiate commonsense conceptions of the world and the greater closeness of the human sciences to common sense. While analytic writers have mostly regarded these features as self-evident, I offer an explanation of them by appealing to Wilhelm Dilthey and the phenomenological tradition. Dilthey suggested that, whereas the natural sciences could individuate their phenomena purely from empirical data, human sciences were compelled to refer to commonsense understanding in order to individuate mental phenomena. I apply this insight to episodes from the history of science, and use it to contrast the work of Alfred Schutz and Carl G. Hempel in philosophy of social and natural science. In the final section, I argue that Dilthey’s framework also offers resources to challenge present-day reductionism about mental phenomena.

Is Death Irreversible?

There are currently two legally established criteria for death: the irreversible cessation of circulation and respiration and the irreversible cessation of neurologic function. Recently, there have been technological developments that could undermine the irreversibility requirement. In this paper, I focus both on whether death should be identified as an irreversible state and on the proper scope of irreversibility in the biological definition of death. In this paper, I tackle the distinction between the commonsense definition of death and the biological definition of death to show that even the commonsense concept of death is specified by biological facts. Resting on this argument, I argue that any definition of death is a posteriori. Thus, irreversibility is part of any definition of death because the actual phenomenon of death is irreversible. In addition, I show that the proper domain of irreversibility in a definition of death is circumscribed by physical possibilities and that irreversibility in the definition of death refers to current possibilities for the reversal of relevant biological processes. I conclude that, despite recent technological advancements, death is still irreversible.

SFCA: A Scalable Formal Concepts Driven Architecture for Multi-Field Knowledge Graph Completion

With the proliferation of Knowledge Graphs (KGs), knowledge graph completion (KGC) has attracted much attention. Previous KGC methods focus on extracting shallow structural information from KGs or in combination with external knowledge, especially in commonsense concepts (generally, commonsense concepts refer to the basic concepts in related fields that are required for various tasks and academic research, for example, in the general domain, “Country” can be considered as a commonsense concept owned by “China”), to predict missing links. However, the technology of extracting commonsense concepts from the limited database is immature, and the scarce commonsense database is also bound to specific verticals (commonsense concepts vary greatly across verticals, verticals refer to a small field subdivided vertically under a large field). Furthermore, most existing KGC models refine performance on public KGs, leading to inapplicability to actual KGs. To address these limitations, we proposed a novel Scalable Formal Concept-driven Architecture (SFCA) to automatically encode factual triples into formal concepts as a superior structural feature, to support rich information to KGE. Specifically, we generate dense formal concepts first, then yield a handful of entity-related formal concepts by sampling and delimiting the appropriate candidate entity range via the filtered formal concepts to improve the inference of KGC. Compared with commonsense concepts, KGC benefits from more valuable information from the formal concepts, and our self-supervision extraction method can be applied to any KGs. Comprehensive experiments on five public datasets demonstrate the effectiveness and scalability of SFCA. Besides, the proposed architecture also achieves the SOTA performance on the industry dataset. This method provides a new idea in the promotion and application of knowledge graphs in AI downstream tasks in general and industrial fields.

Analysis of the Sustainable Driving Effect of Building Energy Consumption on Economic Development Based on the Sustainable Driving Force Model

The environmental problems caused by building energy consumption (BEC) are becoming increasingly prominent. Reducing building energy consumption can not only effectively curb environmental pollution, but also continue to promote economic development. However, there are few reports on reducing building energy consumption to continuously promote economic development. Sustainable driving force (SDF) has been widely mentioned in past research as a common-sense concept, but few systematic studies have been carried out. The main research objectives of this article include two aspects: On the one hand, this study takes the lead in establishing an SDF model and conducting systematic research on SDF. On the other hand, based on the SDF model, an empirical study is conducted on the sustainable driving effect of reducing building energy consumption on economic development in China. The main conclusions are as follows: (1) Research on the SDF model: Based on common examples, the objective reality of SDF has been theoretically proven. Based on the relevant theories of econometrics, this article believes that the “Granger causality test” and the “distributed lag regression model” can be used as basic tools for quantitative research of SDF models. (2) Research on the application of SDF model: From the Granger causality test, it can be seen that the current BEC intensity is the Granger cause of China’s economic development in the first and eighth cycles of the future. According to the results of the distributed lag regression model, reducing BEC intensity has a sustainable driving effect on economic development. Specifically, in the lagged period 1~8, the sustainable driving effect (reverse effect) of BEC on economic development gradually increases at first (coefficient value from 3878.52 to 5163.87), and then gradually decreases (coefficient value from 5163.87 to 783.534). To sum up, the SDF model can provide a reference for the quantitative study of SDF; studying the sustainable driving effect of BEC on economic development can provide a macro reference for the government to formulate “energy conservation and emission reduction” measures.

Sellars, practical reality, and practical truth

AbstractWilfrid Sellars is usually read as claiming that only the unobservable, theoretical objects which science would postulate at the ideal end of inquiry are real. Against this, Willem deVries has suggested that we can develop a notion of practical reality in the context of Sellars's philosophy which would pertain primarily to commonsense objects. I use deVries's suggestion as a foil to clarify Sellars's own commitments about the practical. I show that the notion of practical reality is not necessary to secure the ultimate legitimacy of practical concepts. We can retreat to the less committed notion of practical truth. I argue that the notion of practical truth may not be sufficient to ensure that ultimately legitimate practical concepts are commonsense concepts in Sellars's framework. Nevertheless, deVries's suggestion highlights that Sellars's vision of an ultimate conceptual system may have been richer than is often thought.

SimpleMind: An open-source software environment that adds thinking to deep neural networks.

Deep neural networks (DNNs) detect patterns in data and have shown versatility and strong performance in many computer vision applications. However, DNNs alone are susceptible to obvious mistakes that violate simple, common sense concepts and are limited in their ability to use explicit knowledge to guide their search and decision making. While overall DNN performance metrics may be good, these obvious errors, coupled with a lack of explainability, have prevented widespread adoption for crucial tasks such as medical image analysis. The purpose of this paper is to introduce SimpleMind, an open-source software environment for Cognitive AI focused on medical image understanding. It allows creation of a knowledge base that describes expected characteristics and relationships between image objects in an intuitive human-readable form. The knowledge base can then be applied to an input image to recognize and understand its content. SimpleMind brings thinking to DNNs by: (1) providing methods for reasoning with the knowledge base about image content, such as spatial inferencing and conditional reasoning to check DNN outputs; (2) applying process knowledge, in the form of general-purpose software agents, that are dynamically chained together to accomplish image preprocessing, DNN prediction, and result post-processing, and (3) performing automatic co-optimization of all knowledge base parameters to adapt agents to specific problems. SimpleMind enables reasoning on multiple detected objects to ensure consistency, providing cross-checking between DNN outputs. This machine reasoning improves the reliability and trustworthiness of DNNs through an interpretable model and explainable decisions. Proof-of-principle example applications are provided that demonstrate how SimpleMind supports and improves deep neural networks by embedding them within a Cognitive AI environment.

Fragmented and conflicted: folk beliefs about vision

Many philosophical debates take for granted that there is such a thing as ‘the’ common-sense conception of the phenomenon of interest. Debates about the nature of perception tend to take for granted that there is a single, coherent common-sense conception of vision, consistent with Direct Realism. This conception is often accorded an epistemic default status. We draw on philosophical and psychological literature on naïve theories and belief fragmentation to motivate the hypothesis that untutored common sense encompasses conflicting Direct Realist and Indirect Realist conceptions: there is no such thing as ‘the’ common-sense conception of vision that could enjoy epistemic default status. To examine this hypothesis, a survey administered an agreement rating task with verbal and pictorial stimuli to lay participants. We found many laypeople simultaneously hold conflicting Direct Realist and Indirect Realist beliefs about vision. Against common assumptions, Direct Realist beliefs are not clearly dominant, and consistent adherence to Direct Realism is not the norm, but the exception. These findings are consistent with recent accounts of belief fragmentation. They forcefully challenge common methodology in philosophical debates about the nature of perception and beyond.

Thomas Reid, Common Sense, and Pragmatism

ABSTRACT This paper deals with a less well-known connection between Thomas Reid’s conception of common sense and pragmatism. The paper starts with an exposition of the different principles of common sense one can find in Reid’s writings and a discussion of their epistemic status. The main focus of the paper is on what one may call ‘Reid’s dilemma of common sense’. I argue that Reid’s writings not only present us with a dilemma of common sense but that they also offer a way out of the dilemma, one that is pragmatist in a certain sense. I also discuss the question whether the proposed way out can work.

From common sense to collective practices: a social-ontological commentary on Gramsci’s concept of common sense

In my paper, I start by providing a brief taxonomy of three types of theories that conceptualize common sense. Thereby, I differentiate between positive, negativist, and dualistic/dialectical theories of common sense. By juxtaposing a “good” form of common sense that he calls buon senso and a “bad” form of common sense he calls senso comune Antonio Gramsci falls undoubtedly within the reach of the dualistic/dialectical theories of common sense. In order to scrutinize Gramsci’s idiosyncratic conceptualization of common sense, I turn to Kate Crehan’s interpretation of this concept and argue that Gramsci’s understanding of common sense is better understood as a social-ontological category that is historical, inclusive, relational, and intersectional. In this framework, collectivity is disclosed as the primary mode both of existence and of practice. Ultimately, and in order to demonstrate the current political ramifications of Gramsci’s understanding of common sense, I turn to his figure of the “modern prince.” As I conclude, it is the “bad” form of common sense as senso comune that is most capable of addressing the current demands of an inclusive and radical democracy.

Rehabilitation of full-arch edentulism with fixed or removable dentures retained by root-form dental implants: A systematic review of outcomes and outcome measures used in clinical research in the last 10 years.

To evaluate outcome measures, methods of assessment, and analysis in clinical studies on the rehabilitation of full-arch edentulism with implant-supported fixed or removable prostheses. A protocol-oriented search was conducted in MEDLINE via PubMed, EMBASE, and Cochrane Library (PROSPERO registration CRD42021265179) from 1 January 2011 to 27 July 2021 to identify longitudinal studies reporting on the rehabilitation of edentulous patients with implant-supported prostheses. The primary aim of this study was the collection of all objectively reported outcomes in each study. Secondary aims included the description of the methods of assessment and analytical methods. Risk of bias was applied according to the study design (randomized controlled trial, cohort study, or descriptive pre-post study). Individual study data were extracted into an outcomes matrix. Outcomes were grouped into domains and descriptively analysed. A network diagram was generated to establish relationships between domains. The present review follows the PRISMA statement. The screening and selection processes resulted in the identification of 491 publications, corresponding to 421 different studies (cohorts of patients). Only 24% of the studies reported the use of EQUATOR network guidelines. Implant failure/survival was the most reported outcome (270 studies), but the criterion to determine implant failure and/or survival was frequently not described or was ambiguous. Implant success was much less frequently reported (88) and was based on several heterogenous composite definitions. Marginal bone levels (233 studies), technical complications (158), and clinical outcomes (150), including peri-implant soft tissue and implant stability assessment, were also frequently reported. Patient-reported outcome measures (PROMs) (145) and function-based outcomes (40) were predominantly reported in isolation from other outcomes and most frequently in removable restorations. While quality of life was evaluated using valid instruments, patient satisfaction was evaluated based on a common sense concept of satisfaction. Economic outcomes were under-reported (13). There is great heterogeneity in the criteria to define implant failure or survival and implant success, which prevents the comparison of rates across studies. Even though studies frequently report multiple outcomes, PROMs are usually reported in isolation from other outcomes. It would be valuable to have a set of core outcome variables and standardized methods of measurement for future studies.

Rehabilitation of full-arch edentulism with fixed or removable dentures retained by root-form dental implants: A systematic review of outcomes and outcome measures used in clinical research in the last 10 years.

To evaluate outcome measures, methods of assessment, and analysis in clinical studies on the rehabilitation of full-arch edentulism with implant-supported fixed or removable prostheses. A protocol-oriented search was conducted in MEDLINE via PubMed, EMBASE, and Cochrane Library (PROSPERO registration CRD42021265179) from 1 January 2011 to 27 July 2021 to identify longitudinal studies reporting on the rehabilitation of edentulous patients with implant-supported prostheses. The primary aim of this study was the collection of all objectively reported outcomes in each study. Secondary aims included the description of the methods of assessment and analytical methods. Risk of bias was applied according to the study design (randomized controlled trial, cohort study, or descriptive pre-post study). Individual study data were extracted into an outcomes matrix. Outcomes were grouped into domains and descriptively analysed. A network diagram was generated to establish relationships between domains. The present review follows the PRISMA statement. The screening and selection processes resulted in the identification of 491 publications, corresponding to 421 different studies (cohorts of patients). Only 24% of the studies reported the use of EQUATOR network guidelines. Implant failure/survival was the most reported outcome (270 studies), but the criterion to determine implant failure and/or survival was frequently not described or was ambiguous. Implant success was much less frequently reported (88) and was based on several heterogenous composite definitions. Marginal bone levels (233 studies), technical complications (158), and clinical outcomes (150), including peri-implant soft tissue and implant stability assessment, were also frequently reported. Patient-reported outcome measures (PROMs) (145) and function-based outcomes (40) were predominantly reported in isolation from other outcomes and most frequently in removable restorations. While quality of life was evaluated using valid instruments, patient satisfaction was evaluated based on a common sense concept of satisfaction. Economic outcomes were under-reported (13). There is great heterogeneity in the criteria to define implant failure or survival and implant success, which prevents the comparison of rates across studies. Even though studies frequently report multiple outcomes, PROMs are usually reported in isolation from other outcomes. It would be valuable to have a set of core outcome variables and standardized methods of measurement for future studies.

Externalism, transparency, and diagonal propositions

Boghossian (Philos Perspect 8:33–50, 1994) argued (a) that externalism is incompatible with a transparency thesis according to which we can know a priori whether any two of our occurrent thoughts have the same or distinct content, and (b) that this transparency thesis is integral to our commonsense conception of rationality, which requires the apriority of our logical abilities. Stalnaker (Our knowledge of the internal world, Oxford University Press, Oxford, 2008) offered a detailed compatibilist response to Boghossian. Boghossian (Philos Stud 155:457–465, 2011) criticized this response, and Stalnaker (Philos Stud 155:467–479, 2011) replied. But the outcome of that important discussion remained unclear, partly because it was unclear how to understand the amended transparency thesis at which Stalnaker was gesturing. My aim in this paper is to settle the matter. I wish (i) to clarify the terms of the debate, also by highlighting the relevant commitments of Stalnaker’s two-dimensionalist account, and (ii) to show that his compatibilist response, which appeals to diagonal propositions, is unsuccessful.

Scaling up Predictive Processing to language with Construction Grammar

ABSTRACT Predictive Processing (PP) is an increasingly influential neurocognitive-computational framework. PP research has so far focused predominantly on lower level perceptual, motor, and various psychological phenomena. But PP seems to face a “scale-up challenge”: How can it be extended to conceptual thought, language, and other higher cognitive competencies? Compositionality, arguably a central feature of conceptual thought, cannot easily be accounted for in PP because it is not couched in terms of classical symbol processing. I argue, using the example of language, that there is no strong reason to think that PP cannot be scaled up to higher cognition. I suggest that the tacitly assumed common-sense conception of language as Generative Grammar (“folk linguistics”) and its notion of composition leads to the scale-up concerns. Fodor’s Language of Thought Hypothesis (LOTH) plays the role of a cognitive computational paradigm for folk linguistics. Therefore, we do not take LOTH as facing problems with higher cognition, at least with regard to compositionality. But PP can plausibly play the role of a cognitive-computational paradigm for an alternative conception of language, namely Construction Grammar. If Construction Grammar is a plausible alternative to folk linguistics, then PP is not in a worse position than LOTH.

Application of Euclidean geometry in the assessment of body posture in a sitting position

IntroductionThe analysis of the literature and the results of my research and experience showed the need to search for biomechanical criteria that will make it possible to describe the body shape, especially in a sitting position, and will facilitate the implementation of therapeutic measures.AimApplication of Euclidean geometry in the body posture assessment.Material and methodsThe study uses Euclidean geometry and the concept of common sense to define mutual relationship of the position of the body of the sternum, the sacrum, thoracic kyphosis and lumbar lordosis. The findings are the result of long-time population observations.Results and discussionThe criteria found in the process of my clinical observations and studies allowed me to show close relationship of simultaneous movements and positions of individual parts of the body, i.e. the body of the sternum and the sacrum, as well as thoracic kyphosis and lumbar lordosis in relation to each other. On the basis of Euclidean geometry, this relationship was equated with the shape of a triangle, showing that common relationship between the specified body parts, the so-called ‘common sense,’ could be a method of assessing the body posture in a sitting position.Conclusions(1) The movement of the sternum and sacrum system as well as a thoracic and lumbar spine are biomechanically correlated with each other. (2) Positioning the sternum to the value of the angle α may be a tip on how to control the body posture.

Deep neural network models of visual cortex reveal curvature and real-world size as organizing principles of mid-level representation

Mid-level visual features directly support an array of behaviors; thus, they may be critical for understanding the functional organization of visual cortex. However, attempts at characterizing mid-level features have been hampered by the difficulty of describing these features in words—they exist in an “ineffable valley” between the describable patterns of low-level vision (e.g., edges) and the commonsense concepts of visual cognition (e.g., objects). Here we developed a novel approach to identify interpretable emergent properties of mid-level representations in deep neural network (DNN) models of visual cortex. Using this approach, we examined DNN models that were fit to scene-evoked fMRI responses in category-selective regions of visual cortex—specifically, scene-selective cortex (sceneDNN) and object-selective cortex (objectDNN). Our method uses a semantically-guided image-occlusion procedure to systematically characterize how DNN activations are driven by the classes of objects within a scene. We examined the relationship between mid-level features and several object properties that have previously been associated with response preferences in visual cortex: curvature, real-world size, animacy, naturalness, and spatial stability. We found that while mid-level features appear complex and difficult to describe at a surface level, large-scale computational analyses can reveal a latent underlying relationship to interpretable object properties. Specifically, we found that the mid-level representations of the sceneDNN support a latent preference for objects that are boxy and large in real-world size. In contrast, mid-level representations of the objectDNN support a complementary preference for objects that are curvy and small in real-world size. These effects were robust to variations of model hyperparameters and were reproducible across different DNN models. Our findings show that curvature and real-world size are emergent organizing principles of mid-level visual representation, and they suggest that differences in mid-level feature tuning may be critical for understanding the organization of visual cortex into category-selective patches.

Percepciones de pequeños propietarios y su disposición a la conservación de la naturaleza en áreas privadas de la Cordillera de la Costa en la Región de Los Ríos, Chile

La inclusión de percepciones e intereses de pequeños propietarios privados es fundamental para garantizar a largo plazo el éxito de estrategias de conservación de la naturaleza. Sin embargo, en Chile estas dimensiones no suelen ser incorporadas en la discusión de políticas públicas y han sido incipientemente abordadas en la literatura académica local. A través de una metodología mixta, que incluyó análisis cartográfico, encuestas de percepción y entrevistas a propietarios y actores clave, se analizó como perciben los propietarios de tierra el desarrollo de medidas de conservación y que tan dispuestos están para implementarlas, en dos localidades de la Cordillera de la Costa de la Región de los Ríos. Esta es una zona de alto valor ecológico que alberga especies en peligro de extinción, como la Rana de Mehuin, además de ser un territorio amenazado por el avance de la industria forestal y otras actividades antrópicas. Los resultados muestran que los propietarios presentan miradas disímiles sobre la conservación de la naturaleza, entendiendo ésta desde concepciones generalizadas y de sentido común vinculadas a la protección del bosque principalmente. Asimismo, la mayoría tendería a incorporarse a estrategias de conservación, pero para ello desearían obtener retribuciones económicas por parte del Estado, considerando que las actividades productivas realizadas en la zona son principalmente de subsistencia. Esto supone la necesidad de considerar estos aspectos al momento de desarrollar planes de conservación a pequeña escala, que permitan también aportar al desarrollo económico local de las comunidades involucradas.

KG-BART: Knowledge Graph-Augmented BART for Generative Commonsense Reasoning

Generative commonsense reasoning which aims to empower machines to generate sentences with the capacity of reasoning over a set of concepts is a critical bottleneck for text generation. Even the state-of-the-art pre-trained language generation models struggle at this task and often produce implausible and anomalous sentences. One reason is that they rarely consider incorporating the knowledge graph which can provide rich relational information among the commonsense concepts. To promote the ability of commonsense reasoning for text generation, we propose a novel knowledge graph augmented pre-trained language generation model KG-BART, which encompasses the complex relations of concepts through the knowledge graph and produces more logical and natural sentences as output. Moreover, KG-BART can leverage the graph attention to aggregate the rich concept semantics that enhances the model generalization on unseen concept sets. Experiments on benchmark CommonGen dataset verify the effectiveness of our proposed approach by comparing with several strong pre-trained language generation models, particularly KG-BART outperforms BART by 5.80, 4.60, in terms of BLEU-3, 4. Moreover, we also show that the generated context by our model can work as background scenarios to benefit downstream commonsense QA tasks.