LAVCap improves automated audio captioning by using optimal transport to align audio and visual features and to fuse them before an LLM decoder generates text, reporting SOTA SPIDEr on AudioCaps.
EnCLAP: Combining neural audio codec and audio-text joint embedding for automated audio cap- tioning,
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.MM 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
LAVCap: LLM-based Audio-Visual Captioning using Optimal Transport
LAVCap improves automated audio captioning by using optimal transport to align audio and visual features and to fuse them before an LLM decoder generates text, reporting SOTA SPIDEr on AudioCaps.