specific method · filed under post-training
Concatenated Routing Replay
A replay method that concatenates expert routing from rollout segments spanning multiple checkpoints instead of recomputing it with newer checkpoints.
- source
- 1
- model
- 1
- lab adopt it
- 1
- strongest
- used
How sources treat it
One count per evidence span, weakest treatment to strongest.
used 1
Documented in
Evidence
1 span quoted from the sources, strongest treatment first.
During training, we use concatenated routing-replay: for samples that span multiple checkpoints, we concatenate the expert routing produced at each rollout segment rather than discarding and recomputing the routing information with new checkpoints.
usedsoftware implementationin DeepSeek-V4.1DeepSeek
Filed alongside
Other methods under post-training :: reinforcement learning algorithm.
Group Relative Policy OptimizationReinforcement LearningGroupwise Advantage RedistributionFreezing the MoE Router During Reinforcement LearningIcePopOff-Policy Sequence MaskingReinforcement Learning from Verifiable RewardsRollout Routing ReplayUnbiased KL EstimateAsynchronous Group Relative Policy OptimizationChain-of-Thought Reinforcement LearningDirect Double-Sided Importance SamplingKeep Sampling MaskMixed Reinforcement LearningPivot Reinforcement LearningReinforcement Learning Post-TrainingAbstention TrainingAdvantage ShapingAgentic RL Task MixCISPO with Length-Weighted Leave-One-Out Group-Relative AdvantagesDerived-Latency PenaltyDomain-Specialized RL ExpertsDomain-Specific GRPO TrainingDropping All-Zero-Advantage Groups