93 research outputs found
Generativ AI i användarcentrerad design : En explorativ studie i UX-designers insikter, erfarenheter och framtidsvision
No full text Detta arbete undersöker hur UX-designers resonerar kring användningen av generativ AI i designprocessen. Studien använder ramverket Dubbel Diamanten för att strukturera designprocessen, det möjliggör en organiserad analys av var och hur GenAI används. Genom en kvalitativ ansats med litteraturstudie och semistrukturerade intervjuer framträder en komplex bild där GenAI både stödjer och utmanar traditionellt designarbete. Resultaten visar att GenAI har störst potential i de utforskande, divergerande faserna där tekniken kan automatisera repetitiva uppgifter, generera kreativa idéer och effektivt analysera stora datamängder. I dessa faser frigörs tid för mer strategiska och konceptuella moment. Däremot saknar GenAI den djupa kontextuella förståelse som krävs i de mer konvergerande faserna där mänsklig expertis och förståelse är avgörande. Designers uttrycker också oro för tillförlitlighet, bias och risken att förlita sig för mycket på GenAI, vilket kan leda till minskad kreativitet och sämre kritiskt tänkande. Etiska aspekter kring bias är centrala, vilket understryker behovet av att designers behåller ägandeskap och ansvar för sina lösningar. Studien visar att även om GenAI erbjuder värdefulla verktyg för att effektivisera och inspirera designprocessen, måste designers aktivt behålla sitt kreativa och kritiska omdöme samt ett tydligt ansvar för sina designlösningar.This study explores how UX designers reason about the use of generative AI in the design process. By applying the Double Diamond framework to structure the design process, the study enables an organized analysis of where and how GenAI is used. Through a qualitative approach combining a literature review and semi-structured interviews, a complex picture emerges where GenAI both supports and challenges traditional design work. The results show that GenAI has the greatest potential in the exploratory, divergent phases, where the technology can automate repetitive tasks, generate creative ideas, and efficiently analyze large amounts of data. In these phases, time is freed up for more strategic and conceptual work. However, GenAI lacks the deep contextual understanding required in the more convergent phases, where human expertise and insight are essential. Designers also express concerns about reliability, bias, and the risk of over-reliance on GenAI, which could lead to reduced creativity and weaker critical thinking. Ethical considerations around bias are central, emphasizing the need for designers to maintain ownership and responsibility for their solutions. The study shows that while GenAI offers valuable tools to streamline and inspire the design process, designers must actively retain their creative and critical judgment, as well as clear responsibility for their design outcomes
Generativ AI i användarcentrerad design : En explorativ studie i UX-designers insikter, erfarenheter och framtidsvision
No full text Detta arbete undersöker hur UX-designers resonerar kring användningen av generativ AI i designprocessen. Studien använder ramverket Dubbel Diamanten för att strukturera designprocessen, det möjliggör en organiserad analys av var och hur GenAI används. Genom en kvalitativ ansats med litteraturstudie och semistrukturerade intervjuer framträder en komplex bild där GenAI både stödjer och utmanar traditionellt designarbete. Resultaten visar att GenAI har störst potential i de utforskande, divergerande faserna där tekniken kan automatisera repetitiva uppgifter, generera kreativa idéer och effektivt analysera stora datamängder. I dessa faser frigörs tid för mer strategiska och konceptuella moment. Däremot saknar GenAI den djupa kontextuella förståelse som krävs i de mer konvergerande faserna där mänsklig expertis och förståelse är avgörande. Designers uttrycker också oro för tillförlitlighet, bias och risken att förlita sig för mycket på GenAI, vilket kan leda till minskad kreativitet och sämre kritiskt tänkande. Etiska aspekter kring bias är centrala, vilket understryker behovet av att designers behåller ägandeskap och ansvar för sina lösningar. Studien visar att även om GenAI erbjuder värdefulla verktyg för att effektivisera och inspirera designprocessen, måste designers aktivt behålla sitt kreativa och kritiska omdöme samt ett tydligt ansvar för sina designlösningar.This study explores how UX designers reason about the use of generative AI in the design process. By applying the Double Diamond framework to structure the design process, the study enables an organized analysis of where and how GenAI is used. Through a qualitative approach combining a literature review and semi-structured interviews, a complex picture emerges where GenAI both supports and challenges traditional design work. The results show that GenAI has the greatest potential in the exploratory, divergent phases, where the technology can automate repetitive tasks, generate creative ideas, and efficiently analyze large amounts of data. In these phases, time is freed up for more strategic and conceptual work. However, GenAI lacks the deep contextual understanding required in the more convergent phases, where human expertise and insight are essential. Designers also express concerns about reliability, bias, and the risk of over-reliance on GenAI, which could lead to reduced creativity and weaker critical thinking. Ethical considerations around bias are central, emphasizing the need for designers to maintain ownership and responsibility for their solutions. The study shows that while GenAI offers valuable tools to streamline and inspire the design process, designers must actively retain their creative and critical judgment, as well as clear responsibility for their design outcomes
Unstable Instantons in A-model Localization
No full textWe apply localization techniques to -twisted theories
of vector multiplets on . We derive formulae for -model partition
functions and correlators as integrals along a real contour, as opposed to a
complex one. Using the correlator formula, we successfully recover the unstable
instanton partition function of pure two-dimensional Yang-Mills theory from the
vacuum expectation value of an -model operator.Comment: 29 pages, typos corrected, references adde
Ab initio complex band structure of conjugated polymers: Effects of hydrid density functional theory and GW schemes
No full text09.08.12 meb Author version. Publisher allows this to be adde
Catalytic enantioselective alkene epoxidation using novel spirocyclic N-carbethoxy-azabicyclo[3.2.1]octanones
No full text09.08.12 meb. Accepted author version, publisher allows this to be adde
Environnement textuel et réception du texte médiéval. La deuxième vie de la Chronique de Dalimil
No full textCette analyse propose d’apporter un éclairage nouveau sur la réception de la Chronique de Dalimil grâce à l’analyse des « assemblages textuels » constitutifs des codices qui la renferment. Si l’enquête nous surprend dans un premier temps par l’importance des textes de divertissement au sein du corpus et le recul de la portée politique du texte, ce constat est néanmoins en conformité avec les transformations de la société qui avaient rendu caduc le message de Dalimil. L’écrasante majorité des textes en tchèque rend compte des progrès de la vernacularisation ainsi que du rôle des « transferts culturels » dans l’élaboration de la littérature nationale. La composition des différents codices se fait en outre le reflet de la nouvelle place prise par la littérature dans la société, dans le contexte de l’apparition des « nouveaux publics », tout en illustrant les destinées que pouvait suivre le texte, indépendamment des visées premières de son auteur. Enfin, elle montre que le processus de canonisation de l’œuvre, capable de faire le poids à côté des « hits » de la littérature européenne, était déjà largement engagé à l’orée du xve siècle.This analysis proposes to shed new light on the reception of the Chronicle of Dalimil through the analysis of the “textual collections” constituted by the codices that contain it. While we were initially surprised by the importance of texts made for entertainment within the corpus and the decline in the political scope of the text, this observation is nonetheless in conformity with the transformations of society which had rendered obsolete the message of Dalimil. The overwhelming majority of the Czech texts reflect the progress of vernacularization and the role of “cultural transfers” in the development of national literature. The composition of the different codices is also a reflection of the new place taken by literature in society, in the context of the appearance of the “new publics”, while illustrating the different fates that might befall the text, independently of the aims of its author. Finally, it shows that the process of canonization of the work, capable of competing with the “hits” of European literature, was already very much under way at the beginning of the fifteenth century
Real intersection homology II: A local duality obstruction
No full textWe show that the intersection pairing on our real intersection homology
groups is not a dual pairing in general.Comment: 7 pages. Second author grant support adde
Mesh-Tension Driven Expression-Based Wrinkles for Synthetic Faces
No full textRecent advances in synthesizing realistic faces have shown that synthetic training data can replace real data for various face-related computer vision tasks. A question arises: how important is realism? Is the pursuit of photorealism excessive? In this work, we show otherwise. We boost the realism of our synthetic faces by introducing dynamic skin wrinkles in response to facial expressions, and observe significant performance improvements in downstream computer vision tasks. Previous approaches for producing such wrinkles either required prohibitive artist effort to scale across identities and expressions, or were not capable of reconstructing high-frequency skin details with sufficient fidelity. Our key contribution is an approach that produces realistic wrinkles across a large and diverse population of digital humans. Concretely, we formalize the concept of mesh-tension and use it to aggregate possible wrinkles from high-quality expression scans into albedo and displacement texture maps. At synthesis, we use these maps to produce wrinkles even for expressions not represented in the source scans. Additionally, to provide a more nuanced indicator of model performance under deformations resulting from com-pressed expressions, we introduce the 300W-winks evaluation subset and the Pexels dataset of closed eyes and winks.Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.Pattern Recognition and Bioinformatic
Bibliografiska undersökningar tillägnade Claes Annerstedt på hans sjuttiofemårsdag den 7 juni 1914. Festskrift utg. af Nordisk tidskrift för bok- och biblioteksväsen.
No full textLange, H.O. De hollandske bogauktioner.--Lewenhaupt, E. Upsala universitetsbiblioteks brefsamlingar.--Lundberg, O. Ortnamnsbibliografi.--Lundström, H. Det äldsta tryckta svenska synodaleirkuläret.--Nelson, A. Ett bidrag till Olaus Magnus' bibliografi.--Rudbeck, G. En återfunnen släkttafla.--Walde, O. Henrik Rantzaus bibliotek.--Wieselgren, O. Några anteckningar om våra dagars folklitteratur."Upplagans storlek är tvåhundra numrerade exemplar, hvaraf detta är nr 166."Adde, G. Förteckning öfver Annerstedts utgifna skrifter.--Almquist, J.A. Fögderiindelningen i Sverige 1540-1628 och Kammararkivets landskapshandlingar.--Anderson, A. Bibeln i 1600-talets andaktsböcker.--Carlsson, A.B. En supplik af Gjörwell 1772.--Collijn, I. Bibliotheca "Collegii Societatis Jesu in Suetia".--Dahlgren, E.W. Gamla tyska kartor i Kungl. biblioteket.--Ekholm, G. Upsala universitetsbiblioteks gravyrsamling.--Grape, A. Om bröderna Salan och deras handskriftsamling.--Gödel, V. Erik Sparre.--Mode of access: Internet
LAB: Learnable Activation Binarizer for Binary Neural Networks
No full textBinary Neural Networks (BNNs) are receiving an up-surge of attention for bringing power-hungry deep learning towards edge devices. The traditional wisdom in this space is to employ sign(.) for binarizing feature maps. We argue and illustrate that sign(.) is a uniqueness bottleneck, limiting information propagation throughout the network. To alleviate this, we propose to dispense sign(.), replacing it with a learnable activation binarizer (LAB), allowing the network to learn a fine-grained binarization kernel per layer - as opposed to global thresholding. LAB is a novel universal module that can seamlessly be integrated into existing architectures. To confirm this, we plug it into four seminal BNNs and show a considerable accuracy boost at the cost of tolerable increase in delay and complexity. Finally, we build an end-to-end BNN (coined as LAB-BNN) around LAB, and demonstrate that it achieves competitive performance on par with the state-of-the-art on ImageNet. Our code can be found in our repository: https://github.com/sfalkena/LAB.Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.Pattern Recognition and Bioinformatic
- …
