Open Access Journal of Applied Science and Technology(OAJAST)

Financial Physics Lab, Finland

Publications

• Research Article   
  Quantum Hierarchy for Understanding LLM Representations by Modeling Linear Projections and Nonlinear Dynamics
  Author(s): Timo Aukusti Laine*
  
  Large Language Models (LLMs) excel in natural language tasks, but their high-dimensional embedding spaces pose significant interpretability challenges. Current approaches often linearize these spaces, overlooking the complex dynamics inherent in Transformer architectures. This article proposes a quantum framework to analyze LLM representations, leveraging quantum mechanical tools to explore semantic relationships and contextual influences. We introduce a layered hierarchy of semantic spaces and demonstrate that a classical LLM embedding system has an exact quantum mechanical analogue. Using this analogue, we model phenomena such as the modulation of Semantic Noise, the emergence of hallucinations via quantum tunneling, and the formation of stable semantic representations as soliton solutions. Furthermore, we present a simple quantum circuit design, demonstrating the possibility of usi.. Read More»
  

  Abstract HTML PDF