LLMs discover latent planning strategies up to five steps during training and execute them up to eight steps at test time, with larger models reaching seven under few-shot prompting, revealing a dissociation between discovery and execution.
Transformers struggle to learn to search
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
The Depth Ceiling: On the Limits of Large Language Models in Discovering Latent Planning
LLMs discover latent planning strategies up to five steps during training and execute them up to eight steps at test time, with larger models reaching seven under few-shot prompting, revealing a dissociation between discovery and execution.