o1-like models overthink easy tasks; self-training reduces compute use without accuracy loss on GSM8K, MATH500, GPQA, and AIME.
Chain-of-thought prompting elicits reasoning in large language models
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
Post-training on reasoning tasks sparks the emergence of specialized attention heads that enable structured computation, with SFT adding stable heads while GRPO uses dynamic activation and pruning tied to reward signals, and controllable think models relying on compensatory heads instead of specific
citing papers explorer
-
Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs
o1-like models overthink easy tasks; self-training reduces compute use without accuracy loss on GSM8K, MATH500, GPQA, and AIME.
-
Thinking Sparks!: Emergent Attention Heads in Reasoning Models During Post Training
Post-training on reasoning tasks sparks the emergence of specialized attention heads that enable structured computation, with SFT adding stable heads while GRPO uses dynamic activation and pruning tied to reward signals, and controllable think models relying on compensatory heads instead of specific