Seq1F1B: Efficient Sequence-Level Pipeline Parallelism for Large Language Model Training.
Ao SunWeilin ZhaoXu HanCheng YangZhiyuan LiuChuan ShiMaosong SunPublished in: CoRR (2024)
Keyphrases
- language model
- language modeling
- n gram
- probabilistic model
- query expansion
- document retrieval
- retrieval model
- language modelling
- speech recognition
- test collection
- context sensitive
- document level
- smoothing methods
- information retrieval
- mixture model
- document ranking
- statistical language modeling
- word error rate
- language models for information retrieval
- language model for information retrieval
- dependency structure
- query specific
- relevance model
- pseudo relevance feedback
- cross lingual
- supervised learning
- hidden markov models