RTMC aggregates returns across rollout trees to produce step-level Q-values and advantages, improving pass@1 by 3.2 points over GRPO on SWE-bench Verified.
Ng, Daishi Harada, and Stuart Russell
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
RTMC: Step-Level Credit Assignment via Rollout Trees
RTMC aggregates returns across rollout trees to produce step-level Q-values and advantages, improving pass@1 by 3.2 points over GRPO on SWE-bench Verified.