DualOPSD在模型规模和性能上都有显著提升，是自蒸馏领域的一个新突破。与OPSD相比，DualOPSD在多个基准测试中取得了更好的成绩，值得关注。
DualOPSD是一种不对称交替框架，自适应学生模型和教师模型。在Qwen3-8B上，DualOPSD在AIME 2024、AIME 2025和HMMT 2025上分别比OPSD提升了23.61、13.89和10.00个avg@12点。模型规模越大，准确率提升越明显。所有规模下，DualOPSD都减少了截断，4B规模下的诊断结果还显示教师和学生之间的KL散度更低。
DualOPSD: Adaptive Privileged Teachers for On-Policy Self-Distillation
On-policy self-distillation (OPSD) uses a privileged copy of the student model to provide dense supervision without an external teacher. OPSD keeps this privileged teacher fixed, even though the student distribution and output style change during training. We propose DualOPSD, an asymmetric alternating framework that adapts both policies. The student first learns from the privileged teacher. The teacher then moves toward the updated student distribution on the same student trajectory. This update makes later supervision responsive to the learner and does not require another rollout. On Qwen3-8B in non-thinking mode, DualOPSD improves avg@12 over OPSD by 23.61, 13.89, and 10.00 points on AIME 2024, AIME 2025, and HMMT 2025. Results at 1.7B and 4B show that the accuracy gain depends on model scale. Across all three scales, DualOPSD reduces truncation. The 4B diagnostic also shows lower KL in both directions between the teacher and student.