Transformers Get Stable: An End-to-End Signal Propagation Theory for Language Models.
Akhil KediaMohd Abbas ZaidiSushil KhyaliaJungho JungHarshith GokaHaejun LeePublished in: CoRR (2024)
Keyphrases
- end to end
- language model
- language modeling
- n gram
- document retrieval
- probabilistic model
- speech recognition
- admission control
- information retrieval
- query expansion
- retrieval model
- language modelling
- congestion control
- statistical language models
- multipath
- test collection
- scalable video
- language model for information retrieval
- smoothing methods
- relevance model
- context sensitive
- internet protocol
- pseudo relevance feedback
- translation model
- okapi bm
- ad hoc information retrieval