引用本概念的论文(2)
- DemoPSD:基于分歧调制的策略自蒸馏 DemoPSD: Disagreement-Modulated Policy Self-Distillation arXiv 2607.02502
- 更好的起点,更好的终点:用于压缩推理的自举迭代自推理蒸馏 Better Starts, Better Ends: Bootstrapped Iterative Self-Reasoning Distillation for Compressed Reasoning arXiv 2607.15736