| 1. |
- Aler Tubella, Andrea, 1990-, et al.
(författare)
-
ACROCPoLis : a descriptive framework for making sense of fairness
- 2023
-
Ingår i: FAccT '23. - : ACM Digital Library. - 9781450372527 ; , s. 1014-1025
-
Konferensbidrag (refereegranskat)abstract
- Fairness is central to the ethical and responsible development and use of AI systems, with a large number of frameworks and formal notions of algorithmic fairness being available. However, many of the fairness solutions proposed revolve around technical considerations and not the needs of and consequences for the most impacted communities. We therefore want to take the focus away from definitions and allow for the inclusion of societal and relational aspects to represent how the effects of AI systems impact and are experienced by individuals and social groups. In this paper, we do this by means of proposing the ACROCPoLis framework to represent allocation processes with a modeling emphasis on fairness aspects. The framework provides a shared vocabulary in which the factors relevant to fairness assessments for different situations and procedures are made explicit, as well as their interrelationships. This enables us to compare analogous situations, to highlight the differences in dissimilar situations, and to capture differing interpretations of the same situation by different stakeholders.
|
|
| 2. |
- Björklund, Henrik, et al.
(författare)
-
Implementing a speech-to-text pipeline on the MICO platform
- 2016
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- MICO is an open-source platform for cross-media analysis, querying, and recommendation. It is the major outcome of the European research project Media in Context, and has been contributed to by academic and industrial partners from Germany, Austria, Sweden, Italy, and the UK. A central idea is to group sets of related media objects into multimodal content items, and to process and store these as logical units. The platform is designed to be easy to extend and adapt, and this makes it a useful building block for a diverse set of multimedia applications. To promote the platform and demonstrate its potential, we describe our work on a Kaldi-based speech-recognition pipeline.
|
|
| 3. |
- Björklund, Johanna, 1961-, et al.
(författare)
-
Bridging Perception, Memory, and Inference through Semantic Relations
- 2021
-
Ingår i: Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. - : Association for Computational Linguistics (ACL). - 9781955917094 ; , s. 9136-9142
-
Konferensbidrag (refereegranskat)abstract
- There is a growing consensus that surface form alone does not enable models to learn meaning and gain language understanding. This warrants an interest in hybrid systems that combine the strengths of neural and symbolic methods. We favour triadic systems consisting of neural networks, knowledge bases, and inference engines. The network provides perception, that is, the interface between the system and its environment. The knowledge base provides explicit memory and thus immediate access to established facts. Finally, inference capabilities are provided by the inference engine which reflects on the perception, supported by memory, to reason and discover new facts. In this work, we probe six popular language models for semantic relations and outline a future line of research to study how the constituent subsystems can be jointly realised and integrated.
|
|
| 4. |
- Christopoulos, Arthur, et al.
(författare)
-
THE CONCISE GUIDE TO PHARMACOLOGY 2021/22: G protein-coupled receptors.
- 2021
-
Ingår i: British journal of pharmacology. - : Wiley. - 1476-5381 .- 0007-1188. ; 178 Suppl 1
-
Forskningsöversikt (refereegranskat)abstract
- The Concise Guide to PHARMACOLOGY 2021/22 is the fifth in this series of biennial publications. The Concise Guide provides concise overviews, mostly in tabular format, of the key properties of nearly 1900 human drug targets with an emphasis on selective pharmacology (where available), plus links to the open access knowledgebase source of drug targets and their ligands (www.guidetopharmacology.org), which provides more detailed views of target and ligand properties. Although the Concise Guide constitutes over 500 pages, the material presented is substantially reduced compared to information and links presented on the website. It provides a permanent, citable, point-in-time record that will survive database updates. The full contents of this section can be found at http://onlinelibrary.wiley.com/doi/bph.15538. G protein-coupled receptors are one of the six major pharmacological targets into which the Guide is divided, with the others being: ion channels, nuclear hormone receptors, catalytic receptors, enzymes and transporters. These are presented with nomenclature guidance and summary information on the best available pharmacological tools, alongside key references and suggestions for further reading. The landscape format of the Concise Guide is designed to facilitate comparison of related targets from material contemporary to mid-2021, and supersedes data presented in the 2019/20, 2017/18, 2015/16 and 2013/14 Concise Guides and previous Guides to Receptors and Channels. It is produced in close conjunction with the Nomenclature and Standards Committee of the International Union of Basic and Clinical Pharmacology (NC-IUPHAR), therefore, providing official IUPHAR classification and nomenclature for human drug targets, where appropriate.
|
|
| 5. |
- Dahlgren, Curt, et al.
(författare)
-
Gustafsson, Göran
- 2020
-
Ingår i: The SAGE Encyclopedia of the Sociology of Religion. - 2455 Teller Road, Thousand Oaks, California 91320 : SAGE Publications, Inc.. - 9781473942202 - 9781529714401 ; 1, s. 331-332
-
Bokkapitel (övrigt vetenskapligt/konstnärligt)
|
|
| 6. |
- Dahlgren Lindström, Adam, et al.
(författare)
-
CLEVR-Math : A Dataset for Compositional Language, Visual and Mathematical Reasoning
- 2022
-
Ingår i: Neural-Symbolic Learning and Reasoning 2022. - : Technical University of Aachen. ; , s. 155-170
-
Konferensbidrag (refereegranskat)abstract
- We introduce CLEVR-Math, a multi-modal math word problems dataset consisting of simple math word problems involving addition/subtraction, represented partly by a textual description and partly by an image illustrating the scenario. The text describes actions performed on the scene that is depicted in the image. Since the question posed may not be about the scene in the image, but about the state of the scene before or after the actions are applied, the solver envision or imagine the state changes due to these actions. Solving these word problems requires a combination of language, visual and mathematical reasoning. We apply state-of-the-art neural and neuro-symbolic models for visual question answering on CLEVR-Math and empirically evaluate their performances. Our results show how neither method generalise to chains of operations. We discuss the limitations of the two in addressing the task of multi-modal word problem solving.
|
|
| 7. |
- Dahlgren Lindström, Adam, 1993-
(författare)
-
Learning, reasoning, and compositional generalisation in multimodal language models
- 2024
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- We humans learn language and how to interact with the world through our different senses, grounding our language in what we can see, touch, hear, and smell. We call these streams of information different modalities, and our efficient processing and synthesis of the interactions between different modalities is a cornerstone of our intelligence. Therefore, it is important to study how we can build multimodal language models, where machine learning models learn from more than just text. This is particularly important in the era of large language models (LLMs), where their general capabilities are unclear and unreliable. This thesis investigates learning and reasoning in multimodal language models, and their capabilities to compositionally generalise in visual question answering tasks. Compositional generalisation is the process in which we produce and understand novel sentences, by systematically combining words and sentences to uncover the meaning in language, and has proven a challenge for neural networks. Previously, the literature has focused on compositional generalisation in text-only language models. One of the main contributions of this work is the extensive investigation of text-image language models. The experiments in this thesis compare three neural network-based models, and one neuro-symbolic method, and operationalise language grounding as the ability to reason with relevant functions over object affordances.In order to better understand the capabilities of multimodal models, this thesis introduces CLEVR-Math as a synthetic benchmark of visual mathematical reasoning. The CLEVR-Math dataset involve tasks such as adding and removing objects from 3D scenes based on textual instructions, such as \emph{Remove all blue cubes. How many objects are left?}, and is given as a curriculum of tasks of increasing complexity. The evaluation set of CLEVR-Math includes extensive testing of different functional and object attribute generalisations. We open up the internal representations of these models using a technique called probing, where linear classifiers are trained to recover concepts such as colours or named entities from the internal embeddings of input data. The results show that while models are fairly good at generalisation with attributes (i.e.~solving tasks involving never before seen objects), it is a big challenge to generalise over functions and to learn abstractions such as categories. The results also show that complexity in the training data is a driver of generalisation, where an extended curriculum improves the general performance across tasks and generalisation tests. Furthermore, it is shown that training from scratch versus transfer learning has significant effects on compositional generalisation in models.The results identify several aspects of how current methods can be improved in the future, and highlight general challenges in multimodal language models. A thorough investigation of compositional generalisation suggests that the pre-training of models allow models access to inductive biases that can be useful to solve new tasks. Contrastingly, models trained from scratch show much lower overall performance on the synthetic tasks at hand, but show lower relative generalisation gaps. In the conclusions and outlook, we discuss the implications of these results as well as future research directions.
|
|
| 8. |
- Dahlgren Lindström, Adam, et al.
(författare)
-
Probing Multimodal Embeddings for Linguistic Properties: the Visual-Semantic Case
- 2020
-
Ingår i: Proceedings of the 28th International Conference on Computational Linguistics (COLING). - Stroudsburg, PA, USA : International Committee on Computational Linguistics. ; , s. 730-744
-
Konferensbidrag (refereegranskat)abstract
- Semantic embeddings have advanced the state of the art for countless natural language processing tasks, and various extensions to multimodal domains, such as visual-semantic embeddings, have been proposed. While the power of visual-semantic embeddings comes from the distillation and enrichment of information through machine learning, their inner workings are poorly understood and there is a shortage of analysis tools. To address this problem, we generalize the notion of probing tasks to the visual-semantic case. To this end, we (i) discuss the formalization of probing tasks for embeddings of image-caption pairs, (ii) define three concrete probing tasks within our general framework, (iii) train classifiers to probe for those properties, and (iv) compare various state-of-the-art embeddings under the lens of the proposed probing tasks. Our experiments reveal an up to 12% increase in accuracy on visual-semantic embeddings compared to the corresponding unimodal embeddings, which suggest that the text and image dimensions represented in the former do complement each other
|
|
| 9. |
- Dahlgren, Madelene W., et al.
(författare)
-
Type I Interferons Promote Germinal Centers Through B Cell Intrinsic Signaling and Dendritic Cell Dependent Th1 and Tfh Cell Lineages
- 2022
-
Ingår i: Frontiers in Immunology. - : Frontiers Media SA. - 1664-3224. ; 13
-
Tidskriftsartikel (refereegranskat)abstract
- Type I interferons (IFNs) are essential for antiviral immunity, appear to represent a key component of mRNA vaccine-adjuvanticity, and correlate with severity of systemic autoimmune disease. Relevant to all, type I IFNs can enhance germinal center (GC) B cell responses but underlying signaling pathways are incompletely understood. Here, we demonstrate that a succinct type I IFN response promotes GC formation and associated IgG subclass distribution primarily through signaling in cDCs and B cells. Type I IFN signaling in cDCs, distinct from cDC1, stimulates development of separable Tfh and Th1 cell subsets. However, Th cell-derived IFN-γ induces T-bet expression and IgG2c isotype switching in B cells prior to this bifurcation and has no evident effects once GCs and bona fide Tfh cells developed. This pathway acts in synergy with early B cell-intrinsic type I IFN signaling, which reinforces T-bet expression in B cells and leads to a selective amplification of the IgG2c+ GC B cell response. Despite the strong Th1 polarizing effect of type I IFNs, the Tfh cell subset develops into IL-4 producing cells that control the overall magnitude of the GCs and promote generation of IgG1+ GC B cells. Thus, type I IFNs act on B cells and cDCs to drive GC formation and to coordinate IgG subclass distribution through divergent Th1 and Tfh cell-dependent pathways.
|
|
| 10. |
- Dahlgren, Peter, 1983, et al.
(författare)
-
Reinforcing spirals at work? Mutual influences between selective news exposure and ideological leaning
- 2019
-
Ingår i: European Journal of Communication. - : SAGE Publications. - 0267-3231 .- 1460-3705. ; 34:2, s. 159-174
-
Tidskriftsartikel (refereegranskat)abstract
- The growth of partisan news sources has raised concerns that people will increasingly select attitude-consistent information, which might lead to increasing political polarization. Thus far, there is limited research on the long-term mutual influences between selective exposure and political attitudes. To remedy this, this study investigates the reciprocal influences between selective exposure and political attitudes over several years, using a three-wave panel survey conducted in Sweden during 2014–2016. More specifically, we analyse how ideological selective exposure to both traditional and online news media influences citizens’ ideological leaning. Findings suggest that (1) people seek-out ideologically consistent print news and online news and (2) such attitude-consistent news exposure reinforces citizens’ ideological leaning over time. In practice, however, such reinforcement effects are hampered by (3) relatively low overall ideological selective exposure and a (4) significant degree of cross-cutting news exposure online. These findings are discussed in light of selective exposure theory and the reinforcing spirals model.
|
|
| 11. |
- Dahlgren, Peter, 1983, et al.
(författare)
-
Selective online exposure and political polarization during Swedish election campaigns: a longitudinal analysis using four waves of panel data
- 2016
-
Ingår i: 6th ECREA European Communication Conference, Prague.
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- Internet has made it possible for individuals to increasingly select political news that match their political attitudes. This selective exposure also has the potential to mutually reinforce existing attitudes. However, very little is known about the long-term consequences, especially during election periods. We draw upon the reinforcing spirals model to study the mutual reinforcements between selective exposure and political ideology, by using a four-wave panel during five months with a representative random sample (n=2,281) from Sweden during the 2014 European parliamentary election and Swedish national election. Results suggests that individuals are not becoming more extreme in their political ideology during the election period, regardless of whether they are exposed to attitude-consistent or attitude-inconsistent news content.
|
|
| 12. |
- Flores-Langarica, Adriana, et al.
(författare)
-
Intestinal CD103+CD11b+ cDC2 conventional dendritic cells are required for primary CD4+ T and B cell responses to soluble flagellin
- 2018
-
Ingår i: Frontiers in Immunology. - : Frontiers Media SA. - 1664-3224. ; 9:OCT
-
Tidskriftsartikel (refereegranskat)abstract
- Systemic immunization with soluble flagellin (sFliC) from Salmonella Typhimurium induces mucosal responses, offering potential as an adjuvant platform for vaccines. Moreover, this engagement of mucosal immunity is necessary for optimal systemic immunity, demonstrating an interaction between these two semi-autonomous immune systems. Although TLR5 and CD103+CD11b+ cDC2 contribute to this process, the relationship between these is unclear in the early activation of CD4+ T cells and the development of antigen-specific B cell responses. In this work, we use TLR5-deficient mice and CD11c-cre.Irf4fl/fl mice (which have reduced numbers of cDC2, particularly intestinal CD103+CD11b+ cDCs), to address these points by studying the responses concurrently in the spleen and the mesenteric lymph nodes (MLN). We show that CD103+CD11b+ cDC2 respond rapidly and accumulate in the MLN after immunization with sFliC in a TLR5-dependent manner. Furthermore, we identify that whilst CD103+CD11b+ cDC2 are essential for the induction of primary T and B cell responses in the mucosa, they do not play such a central role for the induction of these responses in the spleen. Additionally, we show the involvement of CD103+CD11b+ cDC2 in the induction of Th2-associated responses. CD11c-cre.Irf4fl/fl mice showed a reduced primary FliC-specific Th2-associated IgG1 responses, but enhanced Th1-associated IgG2c responses. These data expand our current understanding of the mucosal immune responses promoted by sFliC and highlights the potential of this adjuvant for vaccine usage by taking advantage of the functionality of mucosal CD103+CD11b+ cDC2.
|
|
| 13. |
-
HHAI 2024: Hybrid Human AI Systems for the Social Good : Proceedings of the Third International Conference on Hybrid Human-Artificial Intelligence
- 2024
-
Proceedings (redaktörskap) (refereegranskat)abstract
- The field of hybrid human-artificial intelligence (HHAI), although primarily driven by developments in AI, also requires fundamentally new approaches and solutions. Multidisciplinary in nature, it calls for collaboration across various research domains, such as AI, HCI, the cognitive and social sciences, philosophy and ethics, and complex systems, to name but a few. This book presents the proceedings of HHAI 2024, the 3rd International Conference on Hybrid Human-Artificial Intelligence, held from 10-14 June 2024 in Malmö, Sweden. The focus of HHAI 2024 was on artificially-intelligent systems that cooperate synergistically, proactively and purposefully with humans, amplifying rather than replacing human intelligence. A total of 62 submissions were received for the main track of the conference, of which 31 were accepted for presentation after a thorough double blind review process. These comprised 9 full papers, 5 blue sky papers, and 17 working papers, making the final acceptance rate for full papers 29%. Acceptance rate across all tracks of the main program was 50%. This book contains all submissions accepted for the main track, as well as the proposals for the Doctoral Consortium and extended abstracts from the Posters and Demos track. Topics covered include human-AI interaction and collaboration; learning, reasoning and planning with humans and machines in the loop; fair, ethical, responsible, and trustworthy AI; societal awareness of AI; and the role of design and compositionality of AI systems in interpretable/collaborative AI, among others. Providing a current overview of research and development, the book will be of interest to all those working in the field and facilitate the ongoing exchange and development of ideas across a range of disciplines.
|
|
| 14. |
- Holmström, Emma, et al.
(författare)
-
Keeping mixtures of Norway spruce and birch in production forests: insights from survey data
- 2021
-
Ingår i: Scandinavian Journal of Forest Research. - : Informa UK Limited. - 0282-7581 .- 1651-1891. ; 36, s. 155-163
-
Tidskriftsartikel (refereegranskat)abstract
- Admixtures of birch in Norway spruce plantations are being promoted as a means to increase habitat and species diversity. The implications of this mixture were analysed with regional survey data from southern Sweden. Permanent sample plots from the Swedish National Forest Inventory (NFI), with Norway spruce and admixture of birch, were used to describe the temporal trends in the admixture, regarding species composition and competitive strength. Observations from thinned plots show a higher harvest removal in birch (35%) than for Norway spruce (19%). Observations without thinnings in the period before measurement showed that individual birch tree growth was lower compared to Norway spruce and it decreased even more with increasing stand age and competition. In addition, a complementary field survey, with multiple distributed sample plots in each stand, was used to detect within-stand variation of species composition and density. Although within-stand heterogeneity was larger in mixed stands in terms of species composition, it was not different from Norway spruce monocultures in terms of stand density. These two surveys show that the admixture of birch, for several reasons, decreases over stand age and although birch increases tree species diversity, it does not necessary imply a change in density.
|
|
| 15. |
- Methnani, Leila, et al.
(författare)
-
The impact of mixed-initiative on collaboration in hybrid AI
- 2024
-
Ingår i: HHAI 2024: hybrid human AI systems for the social good. - Amsterdam : IOS Press. - 9781643685229 ; , s. 469-471
-
Konferensbidrag (refereegranskat)abstract
- This paper explores the integration of mixed-initiative systems in human-AI teams to improve coordination and communication in Search and Rescue (SAR) scenarios, leveraging dynamic control sharing to enhance operational effectiveness.
|
|
| 16. |
- Westman, Johannes, et al.
(författare)
-
Extracellular Histones Induce Chemokine Production in Whole Blood Ex Vivo and Leukocyte Recruitment In Vivo.
- 2015
-
Ingår i: PLoS Pathogens. - : Public Library of Science (PLoS). - 1553-7366 .- 1553-7374. ; 11:12
-
Tidskriftsartikel (refereegranskat)abstract
- The innate immune system relies to a great deal on the interaction of pattern recognition receptors with pathogen- or damage-associated molecular pattern molecules. Extracellular histones belong to the latter group and their release has been described to contribute to the induction of systemic inflammatory reactions. However, little is known about their functions in the early immune response to an invading pathogen. Here we show that extracellular histones specifically target monocytes in human blood and this evokes the mobilization of the chemotactic chemokines CXCL9 and CXCL10 from these cells. The chemokine induction involves the toll-like receptor 4/myeloid differentiation factor 2 complex on monocytes, and is under the control of interferon-γ. Consequently, subcutaneous challenge with extracellular histones results in elevated levels of CXCL10 in a murine air pouch model and an influx of leukocytes to the site of injection in a TLR4 dependent manner. When analyzing tissue biopsies from patients with necrotizing fasciitis caused by Streptococcus pyogenes, extracellular histone H4 and CXCL10 are immunostained in necrotic, but not healthy tissue. Collectively, these results show for the first time that extracellular histones have an important function as chemoattractants as their local release triggers the recruitment of immune cells to the site of infection.
|
|
| 17. |
- Woldemariam, Yonas Demeke, et al.
(författare)
-
Adapting language specific components of cross-media analysis frameworks to less-resourced languages : the case of Amharic
- 2020
-
Ingår i: Proceedings of the 1st Joint SLTU and CCURL Workshop (SLTU-CCURL 2020). - 9791095546351 ; , s. 298-305
-
Konferensbidrag (refereegranskat)abstract
- We present an ASR based pipeline for Amharic that orchestrates NLP components within a cross media analysis framework (CMAF). One of the major challenges that are inherently associated with CMAFs is effectively addressing multi-lingual issues. As a result, many languages remain under-resourced and fail to leverage out of available media analysis solutions. Although spoken natively by over 22 million people and there is an ever-increasing amount of Amharic multimedia content on the Web, querying them with simple text search is difficult. Searching for, especially audio/video content with simple key words, is even hard as they exist in their raw form. In this study, we introduce a spoken and textual content processing workflow into a CMAF for Amharic. We design an ASR-named entity recognition (NER) pipeline that includes three main components: ASR, a transliterator and NER. We explore various acoustic modeling techniques and develop an OpenNLP-based NER extractor along with a transliterator that interfaces between ASR and NER. The designed ASR-NER pipeline for Amharic promotes the multi-lingual support of CMAFs. Also, the state-of-the art design principles and techniques employed in this study shed light for other less-resourced languages, particularly the Semitic ones.
|
|
