Ανάλυση γλωσσικών μοντέλων για τη δημιουργία ταξονομίας ανά τομέα και τύπο προβλήματος.

Μεθοδολογία: Desk-based research.

Πηγή δεδομένων: huggingface, ollama, llamacpp etc.

Οπτικές ανάλυσης ανά μοντέλο: (per model aspects of analysis)

Δημιουργία πίνακα (γραμμές τα μοντέλα και στήλες τα παρακάτω):

type of size (small, medium, large) | size of the model (# parameters) | implementation mode (cloud or native) | quantisation of the model capability | type of task used (problem each model is created for or mostly used for) | domain of application | type of embeddings | performance or (performance metrics use) | thinking capability | multilinguality | multimodality | coding capabilities | base infrastructure requirement (CPU is ok or GPU is necessary?) | testing or production level | type of model (base model, instruct model, transformer/distilled etc.)

- ΤΝ

- ΠΣ

analysis / desk-based search

literature review

https://link.springer.com/content/pdf/10.1007/s10115-024-02120-8.pdf

Raiaan, M. A. K., Mukta, M. S. H., Fatema, K., Fahad, N. M., Sakib, S., Mim, M. M. J., ... & Azam, S. (2024). A review on large language models: Architectures, applications, taxonomies, open issues and challenges. IEEE access, 12, 26839-26874.

https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=10838047