Alphazero-like Tree-Search can Guide Large Language Model Decoding and Training.
Xidong FengZiyu WanMuning WenYing WenWeinan ZhangJun WangPublished in: CoRR (2023)
Keyphrases
- language model
- tree search
- language modeling
- n gram
- document retrieval
- probabilistic model
- constraint propagation
- information retrieval
- speech recognition
- search algorithm
- mathematical programming
- query expansion
- branch and bound
- test collection
- language modelling
- smoothing methods
- retrieval model
- mixture model
- search tree
- context sensitive
- ad hoc information retrieval
- path finding
- bayesian networks
- machine learning
- state space
- translation model
- high dimensional
- dirichlet prior