Edge-first collaborative inference using small language models with cloud fallback is presented as a viable design, supported by a small experiment in which load-aware scheduling halves cloud offload costs.
(2024, Dec.) Incident report: Service down on december 12, 2024
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.DC 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
Smaller, Smarter, Closer: The Edge of Collaborative Generative AI
Edge-first collaborative inference using small language models with cloud fallback is presented as a viable design, supported by a small experiment in which load-aware scheduling halves cloud offload costs.