Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,657 papers · 148 categories

Trend · papers per month

count% of corpus
1122 · May 202619922001200920172026
3 results for “MeanFlow”
On Variance Reduction in Learning Mean Flowscs.LG

MeanFlow training is unstable due to misusing conditional velocity, leading to variance issues.

problem Unstable training of MeanFlow due to variance problems.
method Theoretical analysis and derivation of optimal coefficient in closed form.
result The optimal coefficient in MeanFlow training minimizes variance but not necessarily quality.
MeanFlowunstable trainingvariance reductionconditional velocityMonte Carlo control variate
2026-05-10abs ↗pdf ↗
On Variance Reduction in Learning Mean Flowscs.LG

This work addresses unstable MeanFlow training by optimizing a coefficient in the loss function.

problem Unstable training of MeanFlow models with non-decreasing loss and unbounded gradient variance.
method Established a theory attributing the instability to misuse of the conditional velocity field, derived the optimal coefficient, and showed practical realizations.
result Optimal coefficient yields up to 54% improvement in sample quality and monotone FID trend.
MeanFlowunstable traininggradient variancecontrol variateJacobi-vector product
2026-05-10abs ↗pdf ↗
What matters for Representation Alignment: Global Information or Spatial Structure?cs.CV

iREPA shows spatial structure, not global semantic, drives generation performance in REPA.

problem Understanding what aspect of the target representation matters for generation.
method Empirical analysis of 27 vision encoders, two modifications to REPA.
result Spatial structure, not global semantic, drives generation performance.
representation alignmentspatial structureglobal informationgenerative modelsconvolutional layers
2025-12-11abs ↗pdf ↗