Skip to main content
Search

# Maximal-Coupling-Routed Teacher Supervision

1 related articles

Subscribe to updates
  1. arXiv

    SAKI routes teacher supervision through maximal-coupling accept/correct events, lifting Mean@8 and Pass@8 for both 1.7B and 0.6B students across seven math reasoning benchmarks

    SAKI realizes a KL-constrained teacher-guided rollout through maximal coupling and reuses the realized accept/correct events as a token-level supervision router: accepted positions keep sampled-token reverse-KL, correction positions switch to direct supervision on the teacher's highest-probability token, and the correction probability is exactly TV(p_t,q_t) so the same trust-region radius upper-bounds intervention frequency; an engine-resident speculative verifier preserves the exact-q trajectory distribution and coupling semantics while improving matched-workload rollout throughput by 4.22x, and across seven mathematical reasoning benchmarks SAKI improves Mean@8 and Pass@8 over the matched teacher-guided baseline for both 1.7B and 0.6B students.

You’re all caught up