View source for Mirasol3B

Mirasol3B is an autoregressive multimodal model for time-aligned video and audio.

{{infobox paper
| name = Mirasol3B
| full_name = Mirasol3B: A Multimodal Autoregressive Model for Time-Aligned and Contextual Modalities
| arxiv_link = https://arxiv.org/abs/2311.05698
| project_page = https://research.google/blog/scaling-multimodal-understanding-to-long-videos/
| twitter_link = https://twitter.com/GoogleAI/status/1724553024088191211
| date = February 2024
| authors = AJ Piergiovanni, Isaac Noble, Dahun Kim, Michael Ryoo, Victor Gomes, Anelia Angelova
}}

[[Category: Papers]]