Na minha lista:
| Autor principal: | |
|---|---|
| Formato: | Recurso digital |
| Idioma: | inglês |
| Publicado em: |
Zenodo
2026
|
| Assuntos: | |
| Acesso em linha: | https://doi.org/10.5281/zenodo.18237804 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
Sumário:
- <p>This paper proposes TELOWAQ, a token-based model that provides a structural framework for analyzing and quantifying personal learning weight patterns in human cognitive processes.</p> <p><br>Existing approaches to learning analysis often rely on qualitative indicators such as time spent, task completion counts, or surface-level behavioral observations, while failing to capture the underlying structural and weighted characteristics of individual cognition and decision-making.</p> <p><br>TELOWAQ addresses this gap by representing learning as a weighted interaction between discrete informational tokens—originating from text-based interaction with large language models—and internal cognitive states. This representation enables structural interpretation of learning dynamics without reducing cognition to outcome-based performance metrics.</p> <p><br>Rather than functioning as a predictive model or a closed algorithmic system, TELOWAQ is presented as an open analytical framework that supports comparative analysis, structural mapping, and cross-domain interpretation of learning behaviors. The framework emphasizes adaptability, interpretability, and extensibility, allowing it to be applied across educational, artificial intelligence, and human–machine interaction contexts.</p> <p><br>By reframing learning as a structurally quantifiable process without imposing normative optimization objectives, this work aims to provide a conceptual bridge between human cognition and machine-based learning systems, offering a foundation for further theoretical expansion and applied research.</p>