Sparse autoencoders on foundation model features produce monosemantic neurons for morphological parts, enabling a pipeline that generates 80K trait annotations across 19K insect images with human-validated biological plausibility.
Then, briefly describe the observable morphological features - such as shape, size, color, texture, or distinct markings - based solely on what is visible in the image
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Automatic Image-Level Morphological Trait Annotation for Organismal Images
Sparse autoencoders on foundation model features produce monosemantic neurons for morphological parts, enabling a pipeline that generates 80K trait annotations across 19K insect images with human-validated biological plausibility.