specific method · filed under evaluation
Rollout-based auditing
Assessing task specification–test alignment by attempting each task multiple times with an agent.
- 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 further assess their alignment through rollout-based auditing. Each task is attempted four times by a coding agent.
usedevaluation onlyin MiMo-V2.6 coding-agent task pipelineXiaomi
Filed alongside
Other methods under evaluation.
Prompt-based output standardizationEvaluation without safety filtersRefusal-suppressed variants for capability estimationavg@kAlmost@1Behavior testing with stubs and edge casesDeadline-bounded rollout evaluationDiverse benchmark validation for quantizationDocumenting evaluation configurationEnd-to-end exploit development evaluationEvaluation equivalence thresholdFixed MTP draft-token acceptance lengthFour-run mean pass@1Full trajectory releaseJoint validation protocolMean@5Model–harness co-designOfficial-release-source baseline restrictionPrompt engineering to elicit answersProtocolQA robustness validationRefusal behavior quantificationSample test and length-constraint filteringSandboxed GPU kernel optimization evaluationSingle-agent and multi-agent evaluation configurations