Binary AI feedback from vision-language models improves dynamic object interactions in text-to-video models more effectively than alignment or dynamics metrics via offline RL fine-tuning.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Improving Dynamic Object Interactions in Text-to-Video Generation with AI Feedback
Binary AI feedback from vision-language models improves dynamic object interactions in text-to-video models more effectively than alignment or dynamics metrics via offline RL fine-tuning.