A concept bottleneck model that discovers class concepts directly from images by quantizing them into nearest common words in CLIP space, matching or beating LLM-generated concept bottlenecks without using LLMs.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
V2C-CBM: Building Concept Bottlenecks with Vision-to-Concept Tokenizer
A concept bottleneck model that discovers class concepts directly from images by quantizing them into nearest common words in CLIP space, matching or beating LLM-generated concept bottlenecks without using LLMs.