specific method · filed under other
Training–Inference Consistency
Align training and inference engines on MoE routing and top-p sampling candidate sets.
Also called train–inference consistency.
- source
- 1
- models
- 2
- 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.
We align the training and inference engines on MoE routing and top-p sampling candidate sets
usedsoftware implementationin MiMo-V2.6 RL infrastructureXiaomi
Filed alongside
Other methods under other.
Removing Git-History Leaks from Benchmark ImagesAgentic WorkflowsAutomating Repetitive Infrastructure WorkClosed-Model Performance with Refusal SubstitutionCode-Driven AnimationEarly Release with Feedback-Driven IterationGenerated Artifact Validation with Linters and Runtime TestsJoint Optimization of Architecture, Cache Precision, and DeploymentPrompt Addendum Prohibiting Direct Online-Solution UsePrompt Modality OrderingRemoving Pattern-Matching-Based Anti-Cheat ChecksRequest Caching in the Browser ToolReward-Hacking Detection SystemScore-versus-Cost Efficiency ComparisonUnique Run Identifiers