First non-vacuous PAC-Bayes certificates for large vision and language models in the 100-example low-shot regime, obtained by reinterpreting model merging as a low-dimensional posterior.
Evolutionary optimization of model merging recipes
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Model Merging is Secretly Certifiable: Non-Vacuous Generalisation Bounds for Low-Shot Learning
First non-vacuous PAC-Bayes certificates for large vision and language models in the 100-example low-shot regime, obtained by reinterpreting model merging as a low-dimensional posterior.