Draft proceedings available
Marion Botella is associate professor in Differential Psychology at Université Paris Cité. After defending her thesis describing how emotions are involved in the artistic creative process, she was postdoctoral researcher at the UCLouvain (Belgium) where she examined the impact of creativity on mood. Since 2013, she is conducting her research within the Applied Psychology and Ergonomic Lab (LaPEA). Her research focus on (1) the creative process in various domains (as art, design, science, ...), (2) the teaching of creativity, (3) the development and construction of scales. Her research often involves mixed methods, both quantitative and qualitative.
The creative process according to psychology and methods to explore it
According to psychology, creativity is the ability to produce ideas that are both original and appropriate (Lubart et al., 2015). Fitting in with this definition, the creative process is then the sequence of thoughts and actions that result in an original and adapted production. This process can thus be described according to a macro approach, detailing the stages that make it up, or by a micro approach, detailing the mechanisms within each stage. In this presentation, we will define creativity and, more specifically, the creative process according to psychology, and then look at the methods used to evaluate or observe it.
Tim Van de Cruys's main research interest is natural language processing, with a particular focus on the unsupervised modeling of meaning, the analysis of multivariate language data within the mathematical framework of tensor algebra, and creative language generation. He is currently an associate professor with the Linguistics Department at the Faculty of Arts, KU Leuven. Previously, he was a CNRS researcher affiliated to the IRIT computer science laboratory in Toulouse. He obtained his PhD from the University of Groningen, and held post-doctoral positions at INRIA in Paris, and the University of Cambridge.
Modeling linguistic creativity for computational literary translation
Literary translation poses unique challenges for computational systems - not only in terms of preserving meaning, but in conveying tone, imagery, and style. Creativity plays a central role, especially when translating texts that resist straightforward alignment. In this talk, I present the ERC project TENACITY, which explores unsupervised models of linguistic creativity using tensor-based semantic representations and neural network architectures. These models do not merely replicate language patterns, but aim to understand and generate language with creative intent. I explore how such models can contribute to the task of literary translation, particularly when dealing with metaphor, ambiguity, or stylistic shifts - offering computational techniques that complement the work of human translators in capturing linguistic nuance.