general family · filed under post-training
Asynchronous reinforcement learning infrastructure
Infrastructure for asynchronous RL; the evidence identifies SLIME as one instance but does not specify a single distinguishing mechanism for this broader label.
Also called asynchronous RL infrastructure.
- sources
- 2
- model
- 1
- lab adopt it
- 1
- strongest
- used
How sources treat it
One count per evidence span, weakest treatment to strongest.
used 2
Documented in
Evidence
2 spans quoted from the sources, strongest treatment first.
we developed slime, a novel asynchronous RL infrastructure that substantially improves training throughput and efficiency, enabling more fine-grained post-training iterations.
usedoptimizationin GLM-5Z.ai
we have engineered a new asynchronous reinforcement learning infrastructure.
usedsoftware implementationin GLM-5Z.ai
Filed alongside
Other methods under post-training :: rollout & RL infrastructure.
Asynchronous reinforcement learningPartial rolloutAsynchronous RL frameworks for large-scale agent scaffolds and environment orchestrationToken-in-token-out (TITO)Co-located RL trainingData SchedulerDecoupled control plane and data planeDecoupling agent rollout into sandbox and worker containerDeficit-corrected schedulingLarge-scale asynchronous RL in synthesized tasksOne-step off-policy asynchronous reinforcement learningPredictive Rollout DispatchSample-grained garbage collectionSeamless Rollout EngineSLIMEToken-granularity persistence of rollout statesToken-level interruptionTool ManagerToolboxAdaptive Rollout ConcurrencyAdaptive Rollout SchedulingAgent LoopAgent-centric rollout executionAsynchronous Agent RL algorithms