trends.barnas.net

method · first seen 9 Oct 2026

learned residual correction for sampling

Post-training correction that predicts and adds endpoint error to coarse-step sampler outputs without changing the pretrained model.

Mentions per week

0119 May11 Aug10 NovPapersCodeDiscussion

PapersCodeDiscussion

Earliest sources

Latest sources

Mentioned together with

rectified flowfew-step diffusion sampling