UNO Arena for Evaluating Sequential Decision-Making Capability of Large Language Models.
Zhanyue QinHaochuan WangDeyuan LiuZiyang SongCunhang FanZhao LvJinlin WuZhen LeiZhiying TuDianhui ChuXiaoyan YuDianbo SuiPublished in: CoRR (2024)
Keyphrases
- language model
- sequential decision making
- language modeling
- decision problems
- reinforcement learning
- interactive dynamic influence diagrams
- influence diagrams
- probabilistic model
- n gram
- document retrieval
- language modelling
- retrieval model
- information retrieval
- statistical language models
- speech recognition
- query expansion
- context sensitive
- pseudo relevance feedback
- smoothing methods
- test collection
- translation model
- decision making
- language models for information retrieval
- vector space model
- machine learning
- expected utility
- temporal difference
- state space
- document ranking
- computational complexity
- utility function
- monte carlo
- dynamic programming