Skip to yearly menu bar Skip to main content


Poster

Test-Time Training with Masked Autoencoders

Yossi Gandelsman · Yu Sun · Xinlei Chen · Alexei Efros

Hall J (level 1) #915

Keywords: [ Masked Auto-Encoder ] [ Computer Vision ] [ Test-Time Training ]


Abstract:

Test-time training adapts to a new test distribution on the fly by optimizing a model for each test input using self-supervision.In this paper, we use masked autoencoders for this one-sample learning problem.Empirically, our simple method improves generalization on many visual benchmarks for distribution shifts.Theoretically, we characterize this improvement in terms of the bias-variance trade-off.

Chat is not available.