implementation detail · filed under evaluation
Sample test and length-constraint filtering
Removing submissions that fail sample test cases or exceed length constraints.
- 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.
we eliminated invalid submissions that failed to pass the provided sample test cases or exceeded the length constraints
usedunclearin DeepSeek-V3.2-SpecialeDeepSeek
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 quantificationRollout-based auditingSandboxed GPU kernel optimization evaluationSingle-agent and multi-agent evaluation configurations