In a stochastic k-ary tree, a two-head transformer learns randomized DFS via policy gradient under depth-wise curriculum, generalizes to deeper trees, and adapts to imbalanced goals via discounting.
A mechanistic analysisofatransformertrainedonasymbolicmulti-stepreasoningtask.arXiv preprint arXiv:2402.11917,
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Supervised fine-tuning lets LLMs linearly encode action validity and state predicates, with broader state-space coverage during training improving world-model recovery.
citing papers explorer
-
Agentic Transformers Provably Learn to Search via Reinforcement Learning
In a stochastic k-ary tree, a two-head transformer learns randomized DFS via policy gradient under depth-wise curriculum, generalizes to deeper trees, and adapts to imbalanced goals via discounting.
-
A Close Look At World Model Recovery In Supervised Fine-Tuned LLM Planners
Supervised fine-tuning lets LLMs linearly encode action validity and state predicates, with broader state-space coverage during training improving world-model recovery.