E³-Agent combines a millisecond fast-path router with an event-driven LLM meta-controller to adaptively manage resources for non-stationary edge generative inference, achieving 65-73% latency reduction versus static baselines in simulator tests.
When ai meets sustainable 6g
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
$E^3$-Agent: An Executable and Evolving Agent for Resource Management of Edge Generative Inference
E³-Agent combines a millisecond fast-path router with an event-driven LLM meta-controller to adaptively manage resources for non-stationary edge generative inference, achieving 65-73% latency reduction versus static baselines in simulator tests.