Towards Fast Multilingual LLM Inference: Speculative Decoding and Specialized Drafters.
Euiin YiTaehyeon KimHongseok JeungDu-Seong ChangSe-Young YunPublished in: CoRR (2024)
Keyphrases
- digital libraries
- decoding process
- general purpose
- neural network
- probabilistic inference
- machine learning
- information retrieval
- bayesian networks
- random fields
- inference process
- language independent
- efficient learning
- multilingual information retrieval
- database
- language resources
- decoding algorithm
- probabilistic reasoning
- inference engine
- cross lingual
- bayesian inference
- machine translation
- document retrieval