Reinforcement learning on synthetic graph-theoretic tasks sharply improves LLM graph reasoning, transferring to larger graphs, new encodings, and real-world tasks.
Llama3 foundation models
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
-
G1: Teaching LLMs to Reason on Graphs with Reinforcement Learning
Reinforcement learning on synthetic graph-theoretic tasks sharply improves LLM graph reasoning, transferring to larger graphs, new encodings, and real-world tasks.