作者位次 #2/5ACL 2026Findings· Findings of ACL
Distilling Long- C o T Reasoning through Collaborative Step-wise Multi-Teacher Decoding
Taewon Yun · Jisu Shin · Jeonghwan Choi · Seunghwan Bang · Hwanjun Song
摘要
Distilling large reasoning models (LRMs) has become essential for making their Long-CoT reasoning capabilities practical, as full-scale inference remains computationally prohibitive. Existing curation-based approaches, which select complete reasoning traces post-hoc, overlook the collaborative po...