specific method · filed under inference & serving
Generate-verify-refine loop
A solution is generated, verified, and refined, with evidence describing its use for mathematical tasks and test-time scaling.
Also called Generate-verify-refine test-time scaling, generate–verify–refine methodology.
- sources
- 2
- models
- 2
- labs adopt it
- 2
- 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.
Following the generate–verify–refine methodology of Shao et al. (2025) and Nemotron-Cascade-2 (Yang et al., 2026)
usedinference servingin Nemotron 3 UltraNVIDIA
Filed alongside
Other methods under inference & serving :: reasoning control.
Configurable reasoning effortChain-of-thought reasoningDefault thinking modeDeployment-time scalar effort controlDisabling reasoning via chat-template configurationInference-time reasoning budget controlInterleaved thinking between tool callsThinking mode selectionCross-turn persistent reasoning historyMaximum thinking effortReasoning parserAlways-on thinking modeclear_thinking chat-template parameterControl-token-enabled thinking modeEffort-dependent exponential token-penalty scheduleMedium-effort reasoning modeParallel-fewest-step samplingQwen3 soft thinking switchTask- and mode-specific sampling parameter recommendationsTest-time compute scalingAdaptive reasoningCapped linear reasoning-token length deductionConfigurable thinking or reasoning modeControllable thinking effort via system message and per-token cost