FuLA, a task-agnostic stitching objective that aligns intermediate features through the frozen end network, is claimed to be a more reliable functional similarity metric than task-based stitching.
Revisiting model stitching to compare neural representations
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Grounding Functional Similarity by Invariance-Aware Model Stitching
FuLA, a task-agnostic stitching objective that aligns intermediate features through the frozen end network, is claimed to be a more reliable functional similarity metric than task-based stitching.