Thinking Collapse in reasoning OPSD is driven by teacher gradients at high-entropy forks; AD-OPSD’s dual-perspective soft gate recovers thinking density and up to +4.1% average accuracy.
arXiv preprint arXiv:2511.12661 , year=
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Diagnosing and Mitigating Thinking Collapse in On-Policy Self-Distillation
Thinking Collapse in reasoning OPSD is driven by teacher gradients at high-entropy forks; AD-OPSD’s dual-perspective soft gate recovers thinking density and up to +4.1% average accuracy.