Fine-Tuning Large Vision-Language Models as Decision-Making Agents via Reinforcement Learning.
Yuexiang ZhaiHao BaiZipeng LinJiayi PanShengbang TongYifei ZhouAlane SuhrSaining XieYann LeCunYi MaSergey LevinePublished in: CoRR (2024)
Keyphrases
- language model
- fine tuning
- decision making
- reinforcement learning
- action selection
- language modeling
- multi agent
- multi agent systems
- n gram
- document retrieval
- probabilistic model
- viable alternative
- language modelling
- information retrieval
- retrieval model
- statistical language models
- speech recognition
- context sensitive
- agent receives
- query expansion
- ad hoc information retrieval
- relevance model
- test collection
- query terms
- language model for information retrieval
- smoothing methods
- fine tuned
- machine learning
- multiple agents
- vector space model
- learning agent
- language models for information retrieval
- document ranking
- term dependencies
- data mining
- translation model
- pseudo relevance feedback
- cross lingual
- okapi bm
- retrieval effectiveness
- learning algorithm