2 ms·Scaling Autoregressive Multimodal Models with Vision, Text, Audio, and Action4 points by georgehill 2y ago