CERLLA uses a GPT-4 semantic parser that learns from environment rollouts to map language instructions to Boolean compositions of pretrained value functions, reaching 92% success on 162 BabyAI tasks with lower sample complexity than a non-compositional baseline.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Compositional Instruction Following with Language Models and Reinforcement Learning
CERLLA uses a GPT-4 semantic parser that learns from environment rollouts to map language instructions to Boolean compositions of pretrained value functions, reaching 92% success on 162 BabyAI tasks with lower sample complexity than a non-compositional baseline.