Paper summary
Paper summary: Chain-of-Thought Reasoning Without Prompting
Chain-of-thought reasoning can appear without any prompting: branching on the top-k first tokens during decoding often reveals reasoning paths, and those paths come with higher answer confidence.
- Greedy decoding hides reasoning that's already there; exploring alternative first tokens uncovers it.
- Confidence in the final answer is higher on paths that contain a chain of thought, which gives a way to pick them.