arXiv subjects

Tarmo Nurmi

Publications and source records attributed to Tarmo Nurmi.

4 recordsLinked to original sources

arXiv2026

ToMigo: Interpretable Design Concept Graphs for Aligning Generative AI with Creative Intent

Generative AI often produces results misaligned with user intentions, for example, resolving ambiguous prompts in unexpected ways. Despite existing approaches to clarify intent, a major challenge remains: understanding and influencing AI's interpretation of user intent through simple, direct inputs requiring no expertise or rigid procedures. We present ToMigo, representing intent as design concept graphs: nodes represent choices of purpose, content, or style, while edges link them with interpretable explanations. Applied to graphic design, ToMigo infers intent from reference images and text. We derived a schema of node types and edges from pre-study data, informing a multimodal large language model to generate graphs aligning nodes externally with user intent and internally toward a unified design goal. This structure enables users to explore AI reasoning and directly manipulate the design concept. In our user studies, ToMigo received high alignment ratings and captured most user intentions well. Users reported greater control and found interactive features-editable graphs, reflective chats, concept-design realignment-useful for evolving and realizing their design ideas.

cs.HC↗

arXiv2025

Node-reconfiguring multilayer networks of human brain function

Functional brain network properties are heavily influenced by how the the network nodes are defined. A common approach uses Regions of Interest (ROIs), i.e., predetermined collections of functional magnetic resonance imaging (fMRI) measurement voxels, as nodes. Their definition is always a compromise, as static ROIs cannot capture the dynamics and temporal reconfigurations of the brain areas. Consequently, the ROIs do not align with the functionally homogeneous regions, which can explain the low functional homogeneity values observed for the ROIs. This is in violation of the underlying homogeneity assumption in functional brain network analysis pipelines, which can cause serious problems such as spurious network structure. We introduce the node-reconfiguring multilayer network model, where nodes represent ROIs with boundaries optimized for high functional homogeneity in each time window. In this representation, network layers correspond to time windows, intralayer links depict functional connectivity between ROIs, and interlayer links quantify the overlap between ROIs on different layers. The ROI optimization approach increases functional homogeneity notably, yielding an over 10-fold increase in the fraction of ROIs with high homogeneity compared to static ROIs from the Brainnetome atlas. The optimized ROIs reorganize non-trivially at short time scales of consecutive time windows and across several windows. The amount of reorganization across time windows is connected to intralayer hubness: ROIs with intermediate levels of reorganization have stronger intralayer links than extremely stable or unstable ROIs. Our results demonstrate that reconfiguring parcellations yield more accurate network models of brain function. This supports the ongoing paradigm shift towards the chronnectome that sees the brain as a set of sources with continuously reconfiguring spatial and connectivity profiles.

q-bio.NC↗

arXiv2023

Subnetwork enumeration algorithms for multilayer networks

To understand the structure of a network, it can be useful to break it down into its constituent pieces. This is the approach taken in a multitude of successful network analysis methods, such as motif analysis. These methods require one to enumerate or sample small connected subgraphs of a network, which can be computationally intractable if naive methods are used. Efficient algorithms exists for both enumeration and uniform sampling of subgraphs, and here we generalize the ESU algorithm for a very general notion of multilayer networks. We show that multilayer network subnetwork enumeration introduces nontrivial complications to the existing algorithm, and present two different generalized algorithms that preserve the desired features of unbiased sampling and trivial parallelization. We evaluate these algorithms in synthetic networks and with real-world data, and show that neither of the algorithms is strictly more efficient but rather the choice depends on the features of the data. Having a general algorithm for finding subnetworks makes advanced multilayer network analysis possible, and enables researchers to apply a variety of methods to previously difficult-to-handle multilayer networks in a variety of domains and across many different types of multilayer networks.

physics.soc-ph↗

arXiv2021

Graphlets in multilayer networks

Representing various networked data as multiplex networks, networks of networks and other multilayer networks can reveal completely new types of structures in these system. We introduce a general and principled graphlet framework for multilayer networks which allows one to break any multilayer network into small multilayered building blocks. These multilayer graphlets can be either analyzed themselves or used to do tasks such as comparing different systems. The method is flexible in terms of multilayer isomorphism, automorphism orbit definition, and the type of multilayer network. We illustrate our method for multiplex networks and show how it can be used to distinguish networks produced with multiple models from each other in an unsupervised way. In addition, we include an automatic way of generating the hundreds of dependency equations between the orbit counts needed to remove redundant orbit counts. The framework introduced here allows one to analyze multilayer networks with versatile semantics, and these methods can thus be used to analyze the structural building blocks of myriad multilayer networks.

physics.soc-ph↗