Crayon: Customized On-Device LLM via Instant Adapter Blending and Edge-Server Hybrid Inference.
Jihwan BangJuntae LeeKyuhong ShimSeunghan YangSimyung ChangPublished in: ACL (1) (2024)
Keyphrases
- low latency
- remote server
- industry standard
- bayesian networks
- edge detection
- probabilistic inference
- edge information
- edge detector
- database
- client server
- inference process
- web server
- website
- hybrid learning
- inference engine
- multiscale
- bayesian inference
- belief networks
- single server
- database systems
- mobile terminals
- central server
- graphical models
- real time