A single Monte Carlo tree of reasoning steps, compared branch-by-branch, can provide multiple signals for updating an LLM agent's skill, cutting evolution cost by 73.2% versus SkillOpt.
Step-level value preference optimization for mathematical reasoning
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.AI 1years
2026 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Branch2Skill: Efficient Skill Evolution Through Reasoning Trees
A single Monte Carlo tree of reasoning steps, compared branch-by-branch, can provide multiple signals for updating an LLM agent's skill, cutting evolution cost by 73.2% versus SkillOpt.