Let Training Guide Selection: Online Synthetic Data Filtering via Real-Anchored Utility
Synthetic data can scale training supervision when real-world data are limited, but noise and distribution mismatch can reduce its value. Existing synthetic data selection methods often emphasize fidelity or diversity rather than the learner's evolving needs. We propose FROST, an online framework that estimates synthet...
Yan-Ran Wu, S. Lakdawala, Renzo Tassara Miller et al.
· 0 citations