Exploring internal representations of self-supervised networks: few-shot learning abilities and comparison with human semantics and recognition of objects.

Front Comput Neurosci · 2025 · PMC12679296 · PMID 41357072

Europe PMC DOI

hedged sentence

—

closest p

—

boldest claim

The sentences

a clear trendno p-value reported

In contrast, the supervised model exhibited a clear trend: the average error rate was higher in the shallower layers and gradually decreased in the deeper layers.

also in 12,254 other papers

Quoted from the open-access full text in Europe PMC under the licence the publisher applied. The sentence is reproduced exactly as published; the emphasis is ours.