HeavySkill frames heavy thinking as a learnable inner skill of parallel reasoning then summarization in LLMs that outperforms Best-of-N sampling and can be scaled with reinforcement learning for better agentic performance.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
HeavySkill: Heavy Thinking as the Inner Skill in Agentic Harness
HeavySkill frames heavy thinking as a learnable inner skill of parallel reasoning then summarization in LLMs that outperforms Best-of-N sampling and can be scaled with reinforcement learning for better agentic performance.