LLaVA-OneVision is the first single open LMM to simultaneously achieve strong performance in single-image, multi-image, and video scenarios with cross-scenario transfer capabilities.
DPO training with AI feedback on videos can further yield significant improvement
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
LLaVA-OneVision: Easy Visual Task Transfer
LLaVA-OneVision is the first single open LMM to simultaneously achieve strong performance in single-image, multi-image, and video scenarios with cross-scenario transfer capabilities.