specific method · filed under post-training
Partial rollout
Splits or interrupts long trajectories across rollout phases, allowing work to continue later and reducing long-tail latency; the evidence does not establish a distinct named variant.
Also called Partial rollout for mixed-task RL, partial rollout for synchronous RL, partial rollout scheme, partial rollouts.
- sources
- 3
- models
- 4
- labs adopt it
- 2
- strongest
- used
How sources treat it
One count per evidence span, weakest treatment to strongest.
Documented in
Evidence
5 spans quoted from the sources, strongest treatment first.
we use partial rollout to keep the running batch saturated: when we collect a training batch, we leave the in-flight sequences interrupted and resume them in the next rollout phase
we have maintained a Data Scheduler that targets a specified training distribution across data sources, together with dynamic sampler and partial rollout
To mitigate the long-tail latency that intensifies in long-horizon tasks, we extend the partial rollout scheme from our synchronous RL framework
and use partial rollouts [118] to reduce tail latency from ultra-long trajectories.
We integrate partial rollout [team2025kimi, fu2025areal] to partition overlong trajectories across steps
Filed alongside
Other methods under post-training :: rollout & RL infrastructure.