Applying and Verifying a Small Language Model on Inexpensive Hardware

Authors

  • Jacob Collier-Tenison
  • Ian Walker
  • Gabriel Ferrer Hendrix College

DOI:

https://doi.org/10.32473/flairs.38.1.138977

Keywords:

small language models, robotics, language translation

Abstract

We applied the Phi3 Mini SLM on a Raspberry Pi 5 in two domains - language translation and robot control. To compensate for hallucinations, we included a feedback loop in both applications. Although Phi3 Mini runs slowly, it suffices to test these proof-of-concept applications. In testing our applications, we found that the feedback loop enabled users to compensate to some extent for the limitations of the SLM.

FLAIRS-38 Logo

Downloads

Published

14-05-2025

How to Cite

Collier-Tenison, J., Walker, I., & Ferrer, G. (2025). Applying and Verifying a Small Language Model on Inexpensive Hardware. The International FLAIRS Conference Proceedings, 38(1). https://doi.org/10.32473/flairs.38.1.138977

Issue

Vol. 38 (2025): Proceedings of FLAIRS-38

Section

Posters

License

Copyright (c) 2025 Jacob Collier-Tenison, Ian Walker, Gabriel Ferrer

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.