| 51. |
|
|
| 52. |
- Björklund, Johanna, et al.
(författare)
-
On the N best problem for hypergraphs
- 2016
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- We propose an algorithm for computing the $N$ best roots of a weighted hypergraph, in which the weight function is given over an idempotent and multiplicatively monotone semiring. We give a set of conditions that ensures that the weight function is well-defined and that solutions exist. Under these conditions, we prove that the proposed algorithm is correct. This generalizes a previous result for weighted tree automata, and in doing so, broadens the practical applications.
|
|
| 53. |
- Björklund, Johanna, 1961-, et al.
(författare)
-
Polynomial Graph Parsing with Non-Structural Reentrancies
-
Annan publikation (övrigt vetenskapligt/konstnärligt)abstract
- Graph-based semantic representations are valuable in natural language processing, where it is often simple and effective to represent linguistic concepts as nodes, and relations as edges between them. Several attempts has been made to find a generative device that is sufficiently powerful to represent languages of semantic graphs, while at the same allowing efficient parsing. We add to this line of work by introducing graph extension grammar, which consists of an algebra over graphs together with a regular tree grammar that generates expressions over the operations of the algebra. Due to the design of the operations, these grammars can generate graphs with non-structural reentrancies, a type of node-sharing that is excessively common in formalisms such as abstract meaning representation, but for which existing devices offer little support. We provide a parsing algorithm for graph extension grammars, which is proved to be correct and run in polynomial time.
|
|
| 54. |
- Björklund, Johanna, et al.
(författare)
-
Simulation relations for pattern matching in directed graphs
- 2013
-
Ingår i: Theoretical Computer Science. - : Elsevier. - 0304-3975 .- 1879-2294. ; 485, s. 1-15
-
Tidskriftsartikel (refereegranskat)abstract
- We consider the problem of finding the occurrences of a pattern tree t in a directed graph g, and propose two algorithms, one for preprocessing and one for searching for t in g. It is assumed that the object graph itself is large and static, and that the pattern tree is small and frequently updated. To model varying abstraction levels in the data, we work with partially ordered alphabets and compute simulation relations rather than equivalence relations. In particular, vertices and edges are labelled with elements from a pair of preorders instead of unstructured alphabets. Under the above assumptions, we obtain a search algorithm that runs in time O(height (t) . vertical bar t vertical bar . vertical bar(V-g(+/-)t/R-g(+/-)t vertical bar(2)) where vertical bar (V-g(+/-)t/R-g(+/-)t)vertical bar is the number of equivalence classes in the coarsest simulation relation R-g(+/-)t on the graph g((+/-))t, the disjoint union of g and t. This means that the size of the object graph only affects the running time of the search algorithm indirectly, because of the groundwork done by the preprocessing routine in time O(k . vertical bar g vertical bar . vertical bar(V-g/R-g)vertical bar(2)), where vertical bar(V-g/R-g) is the number of equivalence classes in the coarsest simulation relation R-g on g, taking k = vertical bar V-g vertical bar(2) in the general case and k = height (g) if g is acyclic.
|
|
| 55. |
- Björklund, Johanna, et al.
(författare)
-
Syntactic methods for topic-independent authorship attribution
- 2017
-
Ingår i: Natural Language Engineering. - : CAMBRIDGE UNIV PRESS. - 1351-3249 .- 1469-8110. ; 23:5, s. 789-806
-
Tidskriftsartikel (refereegranskat)abstract
- The efficacy of syntactic features for topic-independent authorship attribution is evaluated, taking a feature set of frequencies of words and punctuation marks as baseline. The features are 'deep' in the sense that they are derived by parsing the subject texts, in contrast to 'shallow' syntactic features for which a part-of-speech analysis is enough. The experiments are made on two corpora of online texts and one corpus of novels written around the year 1900. The classification tasks include classical closed-world authorship attribution, identification of separate texts among the works of one author, and cross-topic authorship attribution. In the first tasks, the feature sets were fairly evenly matched, but for the last task, the syntax-based feature set outperformed the baseline feature set. These results suggest that, compared to lexical features, syntactic features are more robust to changes in topic.
|
|
| 56. |
- Björklund, Johanna, 1961-
(författare)
-
The impact of state merging on predictive accuracy in probabilistic tree automata : Dietze’s conjecture revisited
- 2023
-
Ingår i: Fundamentals of computation theory. - : Springer Nature. - 9783031435867 - 9783031435874 ; , s. 74-87
-
Konferensbidrag (refereegranskat)abstract
- Dietze’s conjecture concerns the problem of equipping a tree automaton M with weights to make it probabilistic, in such a way that the resulting automaton N predicts a given corpus C as accurately as possible. The conjecture states that the accuracy cannot increase if the states in M are merged with respect to an equivalence relation ∼ so that the result is a smaller automaton M∼. Put differently, merging states can never improve predictions. This is under the assumption that both M and M∼ are bottom-up deterministic and accept every tree in C. We prove that the conjecture holds, using a construction that turns any probabilistic version N∼ of M∼ into a probabilistic version N of M, such that N assigns at least as great a weight to each tree in C as N∼ does.
|
|
| 57. |
- Björklund, Johanna, 1961-, et al.
(författare)
-
Towards Semantic Representations with a Temporal Dimension
- 2020
-
Konferensbidrag (refereegranskat)abstract
- We outline the initial ideas for a representational framework for capturing temporal aspects in semantic parsing of multimodal data.As a starting point, we take the Abstract Meaning Representations of Banarescu et al. andpropose a way of extending them to coversequential progressions of events. The firstmodality to be considered is text, but the long-term goal is to also incorporate informationfrom visual and audio modalities, as well ascontextual information.
|
|
| 58. |
- Björklund, Johanna, 1961-
(författare)
-
Tree-to-Graph Transductions with Scope
- 2018
-
Ingår i: Developments in Language Theory. DLT 2018.. - Cham : Springer. - 9783319986531 - 9783319986548 ; , s. 133-144
-
Konferensbidrag (refereegranskat)abstract
- High-level natural language processing requires formal languages to represent semantic information. A recent addition of this kind is abstract meaning representations. These are graphs in which nodes encode concepts and edges relations. Node-sharing is common, and cycles occur. We show that the required structures can be generated through the combination of (i) a regular tree grammar, (ii) a sequence of linear top-down tree transducers, and (iii) a fold operator that merges selected nodes. Delimiting the application of the fold operator to connected subgraphs gains expressive power, while keeping the complexity of the associated membership problem in polynomial time.
|
|
| 59. |
- Björklund, Johanna, et al.
(författare)
-
Z-Automata for Compact and Direct Representation of Unranked Tree Languages
- 2019
-
Ingår i: Implementation and Application of Automata. - Cham : Springer. - 9783030236786 - 9783030236793 ; , s. 83-94
-
Konferensbidrag (refereegranskat)abstract
- Unranked tree languages are valuable in natural language processing for modelling dependency trees. We introduce a new type of automaton for unranked tree languages, called Z-automaton, that is tailored for this particular application. The Z-automaton offers a compact form of representation, and unlike the closely related notion of stepwise automata, does not require a binary encoding of its input. We establish an arc-factored normal form, and prove the membership problem of Z-automata in normal form to be in O(mn), where m is the size of the transition table of the Z-automaton and n is the size of the input tree.
|
|
| 60. |
- Björklund, Mats, et al.
(författare)
-
A demo-genetic analysis of a recovering population of otters in central Sweden
- 2008
-
Ingår i: Animal Conservation. - : Wiley. - 1367-9430 .- 1469-1795. ; 11:6, s. 529-534
-
Tidskriftsartikel (refereegranskat)abstract
- We performed a demo-genetic population viability analysis on a recovering population of otters Lutra lutra in Central Sweden, using data on population size, survival and genetic data from microsatellites. Population data were obtained from genotyping faeces. At present, the size and genetic variability of the population is increasing. We found that survival to first reproduction was the most crucial demographic parameter, and that even slight changes downward in this parameter, might lead to a declining population trajectory. Human factors that can affect mortality are traffic, fishing equipment and traps, and we argue that efforts to minimize road kills by means of safe passages as well as careful fishing efforts in streams and lakes would reduce the risk of extinction. In general, even though the population is now growing and has no inbreeding problem, its small abundance could make it vulnerable to chance events and environmental perturbations.
|
|
| 61. |
- Björklund, Niklas, et al.
(författare)
-
Assessing the confidence in pest freedom gained in the past pine wood nematode surveys
- 2023
-
Ingår i: EFSA supporting publications. - 2397-8325. ; 20
-
Tidskriftsartikel (refereegranskat)abstract
- All Member States of the European Union must conduct annual surveys of pine wood nematode (PWN) to ensure its timely detection. However, the statistical confidence of these surveys is rarely assessed. To facilitate such assessments, we developed two easy-to-use web applications: NoBaSURV-PWN for assessment of the statistical confidence of past PWN surveys, and NoBa Land Cover Retriever for retrieving the land cover data needed in the assessments. This report explains how the statistical confidence of past PWN surveys can be assessed with NoBaSURV-PWN. In addition, the reportpresents the assessments done with the NoBaSURV-PWN applicationfor Estonia, Finland, Lithuania, Norway, and Sweden. The technical details of the developed applications are presented,and some matters that have general relevance for statistical assessment and planning of quarantine pest surveys are discussed. Also, the capacity building activitiesdone in the project are described and their impact is evaluated.The assessments forthe five Nordic-Baltic countries show that, in most of the countries, PWN surveys have been extensive enough to provide evidence for facilitating tradewith a rather high confidence. Yet, the surveys have clearly not been extensive enough to ensure detection of PWN invasions at such an early stage that they could be eradicated.
|
|
| 62. |
|
|
| 63. |
|
|
| 64. |
|
|
| 65. |
|
|
| 66. |
|
|
| 67. |
|
|
| 68. |
|
|
| 69. |
|
|
| 70. |
|
|
| 71. |
|
|
| 72. |
|
|
| 73. |
|
|
| 74. |
|
|
| 75. |
|
|
| 76. |
|
|
| 77. |
|
|
| 78. |
|
|
| 79. |
|
|
| 80. |
|
|
| 81. |
- Björklund, Niklas, et al.
(författare)
-
Rapid Pest Risk Analysis Xyleborinus attenuatus
- 2017
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- This rapid pest risk analysis (PRA) provides a quick assessment of the risks posed by the pest to Sweden, which is the PRA area being assessed. The format is an adapted version of the EPPO Express PRA scheme (EPPO 2012). Definition of terms used as well as the rating scheme and assessments are done in line with the guidance given in the EPPO CAPRA system (EPPO 2011). The likelihood of entry and establishment are assessed considering the current phytosanitary regulation in place with respect to the EU legislation (Council Directive 2000/29/EC). The definition of a quarantine pest follows the regulation (EU) 2016/2031.
|
|
| 82. |
|
|
| 83. |
|
|
| 84. |
|
|
| 85. |
|
|
| 86. |
|
|
| 87. |
|
|
| 88. |
|
|
| 89. |
|
|
| 90. |
|
|
| 91. |
|
|
| 92. |
|
|
| 93. |
|
|
| 94. |
|
|
| 95. |
|
|
| 96. |
|
|
| 97. |
|
|
| 98. |
|
|
| 99. |
- 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.
|
|
| 100. |
- 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
|
|
