Chain-of-Thought (CoT) reasoning is a prompting technique where language models generate intermediate reasoning steps before producing a final answer. CoT prompting dramatically improves performance on complex reasoning tasks and underpins modern reasoning models like those in qwen3. It is foundational to techniques like star (Self-Taught Reasoner) and plays...
- Related
- References
Chain of Thought
Chain-of-Thought (CoT) reasoning is a prompting technique where language models generate intermediate reasoning steps before producing a final answer. CoT prompting dramatically improves performance on complex reasoning tasks and underpins modern reasoning models like those in qwen3. It is foundational to techniques like star (Self-Taught Reasoner) and plays a central role in reasoning-distillation.
Related
- qwen3 — Qwen3 supports thinking/non-thinking mode toggling, leveraging CoT reasoning
- star — Self-Taught Reasoner generates its own CoT traces for self-improvement
- reasoning-distillation — Transferring CoT reasoning abilities from teacher to student models
- thinking-budget — Compute allocation for extended CoT reasoning
References
- Wei et al. (2022). "Chain-of-Thought Prompting Elicits Reasoning in Large Language Models." NeurIPS.