SparseJEPA applies an oi-VAE style sparsity penalty to JEPA and reports modest linear-probe gains, but its theoretical justification is flawed.
Ainsworth, Nicholas J
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
SparseJEPA: Sparse Representation Learning of Joint Embedding Predictive Architectures
SparseJEPA applies an oi-VAE style sparsity penalty to JEPA and reports modest linear-probe gains, but its theoretical justification is flawed.