A new Minecraft crafting benchmark shows current LLM agents solve only a minority of planning tasks, and vision-language models almost never succeed on image observations.
Exploring the capacity of pretrained language models for reasoning about actions and change
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Plancraft: an evaluation dataset for planning with LLM agents
A new Minecraft crafting benchmark shows current LLM agents solve only a minority of planning tasks, and vision-language models almost never succeed on image observations.