Using contrastive examples with vision-language models and a new CLIP-based scoring method called CSP produces more faithful and granular neuron labels than prior activation-only approaches.
In: International Conference on Learning Representations (2021)
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
UNBOX recovers interpretable text concepts that maximally activate classes in black-box vision models by recasting activation maximization as semantic search with LLMs and diffusion models.
citing papers explorer
-
Contrastive Semantic Projection: Faithful Neuron Labeling with Contrastive Examples
Using contrastive examples with vision-language models and a new CLIP-based scoring method called CSP produces more faithful and granular neuron labels than prior activation-only approaches.
-
UNBOX: Unveiling Black-box visual models with Natural-language
UNBOX recovers interpretable text concepts that maximally activate classes in black-box vision models by recasting activation maximization as semantic search with LLMs and diffusion models.