MMAR: Towards Lossless Multi-Modal Auto-Regressive Probabilistic Modeling

Open paper

Benchmark	Model	Rank	Results
visual-question-answering-on-mm-vet	MMAR-0.5B	–	GPT-4 score: 18.49
visual-question-answering-on-mm-vet	MMAR-7B	–	GPT-4 score: 27.80