The Problem of Semantic Gaps in the Context of World Models and Sign-Symbolic Communication

Authors

Chuganskaya A.

Annotation

The current development of artificial intelligence technologies involves many aspects of incorporating tools for analyzing and generating content related to social phenomena and communication. To develop a model that takes into ac-count the significant role of semantic gaps in the study of psychological phenomena and the formalization of social phenomena, we turned to the consideration of this problem in computer science, linguistics, psychology, and applied researches. The task of overcoming the semantic gap has a history within the field of humanities. It does not have a complete solution, which is related to the problem of formalizing knowledge. However, the theories of gaps and se-mantic processing of perceived images in psychology provide directions for building new models that include the development of tools for overcoming the semantic gap. The difficulty to explain the meaning in analyzing of social interactions is connected with the problem of the formation of the volume of the analyzed details of the picture and the need to correlate with the existing experience of the person. In machine learning such visual analyses and constructing relevant prediction of images with social actions demand including semantic level. One of the models is based on the Vector Semiotic Architecture. Another model is a Sign-Based World Model cognitive architecture which make an attempt to address the symbol grounding problem. This model is a complex architecture that takes into account both the level of building images of the objective world based on perception, and the complex images of the world based on reasoning. Sign-symbolic communication as a systemic way of exchanging information with a high degree of semantic depth and the presence of diverse semantic gaps allows us to examine the components of the formation of such semantic gaps.

External links

DOI: 10.1007/978-3-032-39382-1_7

Download the article (PDF) from the conference website: https://icpr2026orgteam.github.io/WSpapers/pdfs/paper_0054.pdf

ResearchGate: https://www.researchgate.net/publication/415064315_The_Problem_of_Semantic_Gaps_in_the_Context_of_World_Models_and_Sign-Symbolic_Communication

Reference link

Chuganskaya, A. (2027). The Problem of Semantic Gaps in the Context of World Models and Sign-Symbolic Communication // In: Sintorn, IM., Pal, U., Lu, S. (eds) Pattern Recognition. ICPR 2026 International Workshops. ICPR 2026. Lecture Notes in Computer Science, vol 17111, pp. 94–104. https://doi.org/10.1007/978-3-032-39382-1_7