Model techniques map
Techniquespost-trainingrollout & RL infrastructure

specific method · filed under post-training

Per-dataset rollout concurrency limiting

Limits dispatcher concurrency separately by dataset to regulate each dataset's share of incoming samples.

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.

the dispatcher can limit concurrency on a per-dataset basis, which helps regulate the proportion of each dataset in the steady-state training batch, indirectly mitigating the length skew by controlling the sources of incoming samples

usedpost trainingin DeepSeek-V4.1DeepSeek

Filed alongside

Other methods under post-training :: rollout & RL infrastructure.