Vision as a Dialect: Unifying Visual Understanding and Generation via Text-Aligned Representations
Paper
• 2506.18898 • Published
• 34
Tar
[NeurIPS 2025] Vision as a Dialect: Unifying Visual Understanding and Generation via Text-Aligned Representations
-
-
Tar
🚀48Unified MLLM with Text-Aligned Representations
-
Tar
🚀3Unified MLLM with Text-Aligned Representations
-
Tar
🚀60Unified MLLM with Text-Aligned Representations
-
ByteDance-Seed/Tar-1.5B
Any-to-Any • 3B • Updated • 53 • 21 -
ByteDance-Seed/Tar-7B
Any-to-Any • 9B • Updated • 66 • 39 -
ByteDance-Seed/Tar-TA-Tok
Updated • 6 -
csuhan/Tar-SANA-600M-512px
Text-to-Image • Updated • 1 -
csuhan/Tar-SANA-600M-1024px
Text-to-Image • Updated -
csuhan/Tar-Lumina2
Updated -
csuhan/tar_1.5B_pretrain_demo
3B • Updated • 1 • 1