An inference-time text classifier guides a frozen captioner's context vectors so generated audio captions emphasize audible actions, improving zero-shot captioning metrics.
Bottom-up and top-down attention for image captioning and visual question answering,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Classifier-Guided Captioning Across Modalities
An inference-time text classifier guides a frozen captioner's context vectors so generated audio captions emphasize audible actions, improving zero-shot captioning metrics.