Selectively reducing the number of tools presented to an LLM, using embedding similarity over individual tools or clusters, improves function-calling success and efficiency on edge devices.
Single-path nas: Designing hardware-efficient convnets in less than 4 hours,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.PF 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Less is More: Optimizing Function Calling for LLM Execution on Edge Devices
Selectively reducing the number of tools presented to an LLM, using embedding similarity over individual tools or clusters, improves function-calling success and efficiency on edge devices.