On Transferring Transferability: Towards a Theory for Size Generalization

NeurIPSSpotlight2025

Authors
Eitan Levin, Yuxin Ma, Mateo Diaz Diaz, Soledad Villar
Venue
NeurIPS 2025
Track
Spotlight

TL;DR

Many modern learning tasks require models that can take inputs of varying sizes…

Opening excerpt from the authors’ abstract. source

Read the paper

Topics

generalization theory

Related NeurIPS papers

← All NeurIPS 2025 Spotlight papers · Browse the whole archive