Why Can GPT Learn In-Context? Language Models Secretly Perform Gradient Descent as Meta-Optimizers.
Damai DaiYutao SunLi DongYaru HaoShuming MaZhifang SuiFuru WeiPublished in: ACL (Findings) (2023)
Keyphrases
- language model
- context sensitive
- language modeling
- document retrieval
- probabilistic model
- retrieval model
- n gram
- test collection
- speech recognition
- query expansion
- statistical language models
- language modelling
- relevance model
- information retrieval
- vector space model
- wordnet
- word error rate
- machine learning
- okapi bm
- ad hoc information retrieval
- language models for information retrieval
- language model for information retrieval