STARLING: Self-supervised Training of Text-based Reinforcement Learning Agent with Large Language Models.
Shreyas BasavatiaKeerthiram MurugesanShivam RatnakarPublished in: ACL (Findings) (2024)
Keyphrases
- language model
- learning agent
- reinforcement learning
- language modeling
- probabilistic model
- n gram
- retrieval model
- information retrieval
- language models for information retrieval
- test collection
- state space
- solving problems
- query expansion
- learning algorithm
- smoothing methods
- training set
- learning tasks
- online learning
- multimedia
- reinforcement learning algorithms
- learning capabilities
- collaborative filtering
- image classification
- e learning
- neural network