MM1 models achieve state-of-the-art few-shot multimodal results by pre-training on a careful mix of image-caption, interleaved, and text-only data with optimized image encoders.
In: EMNLP (2013)
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
-
MM1: Methods, Analysis & Insights from Multimodal LLM Pre-training
MM1 models achieve state-of-the-art few-shot multimodal results by pre-training on a careful mix of image-caption, interleaved, and text-only data with optimized image encoders.