Thinking Leakage: A Causal Audit of NoThink Post-Training in Hybrid Reasoning Models
Paper • 2609.28682 • Published • 1
Zehao Liu
zml5418
AI & ML interests
None yet
Recent Activity
upvoted a paper about 17 hours ago
Thinking Leakage: A Causal Audit of NoThink Post-Training in Hybrid Reasoning Models authored a paper about 17 hours ago
Restoring the Sweet Spot: Pass-Rate Weighted Self-Distillation for LLM Reasoning authored a paper about 17 hours ago
Thinking Leakage: A Causal Audit of NoThink Post-Training in Hybrid Reasoning ModelsOrganizations
None yet
upvoted a paper about 17 hours ago