← back to paper

Review history

arxiv: 2608.03092 · 2 revisions

SMOPD: Multi-Reward Reinforcement Learning via Specialize-and-Merge Online Policy Distillation

1. 2026-08-15 CONDITIONAL MODERATE v1.4.0-alltime-deepseek-medium novelty 6.0
   158776 ms 22149 in 21648 out 2026-08-15T14:53:12.732495+00:00
2. 2026-08-08 CONDITIONAL HIGH v1.4.0-alltime-deepseek-medium novelty 6.0
   169641 ms 22143 in 18320 out 2026-08-08T00:56:25.501955+00:00