Distillation remains essential when RL pass rate is zero
A debate on X over RL versus distillation concluded that distillation remains critical: when a model's pass rate is zero, RL cannot proceed, with practitioners suggesting seeded-then-removed prompts for SFT.
2026-09-25 ~ 2026-09-25 · 2 related posts
- 'If Your Pass Rate Is 0, You Can't RL': Why Distillation Still Matters — fleetwood___ · 2026-09-25
- Zero pass rate blocks RL: use prompt hints then strip them for SFT distillation — xeophon · 2026-09-25