GEXIA expands a single-grained video-text dataset into multiple granularities and aligns them with granularity-dependent iterative attention, achieving strong long-form video understanding results.
Blinkdl/rwkv-lm
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
GEXIA: Granularity Expansion and Iterative Approximation for Scalable Multi-grained Video-language Learning
GEXIA expands a single-grained video-text dataset into multiple granularities and aligns them with granularity-dependent iterative attention, achieving strong long-form video understanding results.