Style or Signature? Artist-Disjoint Evaluation of Style Classification in Frozen Vision Embeddings
Frozen image embeddings from models such as CLIP are increasingly used to classify paintings by art-historical style, with high reported accuracy. We ask whether this accuracy reflects an understanding of style or the recognition of individual artists. Standard evaluation uses random splits in which works by the same artist appear on both sides, so a classifier can succeed by recognising the painter rather than the movement. We re-evaluate style classification under an artist-disjoint protocol, holding out every artist in turn so that no work is ever classified using other works by its own pai
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- FuzzySimilar title/name (fuzzy) · 59%VioletVision-3B →
“Fuzzy title match (0.73): “Style or Signature? Artist-Disjoint Evaluation of Style Clas” ≈ “VioletVision-3B””
- PossiblePossibly related (embedding) · 47%What Makes AI Art Worth Collecting? →
- FuzzySimilar title/name (fuzzy) · 84%pytorch/vision →
“Fuzzy title match (0.92): “Style or Signature? Artist-Disjoint Evaluation of Style Clas” ≈ “pytorch/vision””
- LinkedLinked via arxiv author · 85%Rory Ashton →
“Style or Signature? Artist-Disjoint Evaluation of Style Classification in Frozen Vision Embeddings”
- PossiblePossibly related (embedding) · 46%Machine learning of artistic fingerprints in jazz →