Inference-Time Policy Adapters (IPA): Tailoring Extreme-Scale LMs without Fine-tuning.
Ximing LuFaeze BrahmanPeter WestJaehun JungKhyathi ChanduAbhilasha RavichanderLianhui QinPrithviraj AmmanabroluLiwei JiangSahana RamnathNouha DziriJillian FisherBill Yuchen LinSkyler HallinanXiang RenSean WelleckYejin ChoiPublished in: CoRR (2023)
Keyphrases
- fine tuning
- input output
- fine tune
- viable alternative
- learning management systems
- fine tuned
- operating system
- bayesian inference
- neural network
- bayesian networks
- scale space
- optimal policy
- probabilistic inference
- markov decision process
- inference process
- learning objects
- real time
- random fields
- probabilistic model
- e learning