PaDT Multi-Modal Model series based on Patch-as-Decodable-Token framework. PaDT-MLLM/PaDT_Pro_3B Any-to-Any • 4B • Updated Oct 10 • 319 • 2 PaDT-MLLM/PaDT_Pro_7B Any-to-Any • 8B • Updated Oct 10 • 63 • 2 PaDT-MLLM/PaDT_REC_3B Any-to-Any • 4B • Updated Oct 10 • 16 PaDT-MLLM/PaDT_REC_7B Any-to-Any • 8B • Updated Oct 10 • 9
PaDT-Paper Patch-as-Decodable-Token: Towards Unified Multi-Modal Vision Tasks in MLLMs Paper • 2510.01954 • Published Oct 2 • 12
Patch-as-Decodable-Token: Towards Unified Multi-Modal Vision Tasks in MLLMs Paper • 2510.01954 • Published Oct 2 • 12
PaDT-Dataset Preprocessed datasets used to train PaDT framework. PaDT-MLLM/ReferringImageCaptioning Viewer • Updated Oct 10 • 575k • 282 • 3 PaDT-MLLM/COCO Viewer • Updated Oct 10 • 123k • 434 • 1 PaDT-MLLM/RefCOCO Viewer • Updated Oct 10 • 357k • 463 • 3
PaDT Multi-Modal Model series based on Patch-as-Decodable-Token framework. PaDT-MLLM/PaDT_Pro_3B Any-to-Any • 4B • Updated Oct 10 • 319 • 2 PaDT-MLLM/PaDT_Pro_7B Any-to-Any • 8B • Updated Oct 10 • 63 • 2 PaDT-MLLM/PaDT_REC_3B Any-to-Any • 4B • Updated Oct 10 • 16 PaDT-MLLM/PaDT_REC_7B Any-to-Any • 8B • Updated Oct 10 • 9
PaDT-Dataset Preprocessed datasets used to train PaDT framework. PaDT-MLLM/ReferringImageCaptioning Viewer • Updated Oct 10 • 575k • 282 • 3 PaDT-MLLM/COCO Viewer • Updated Oct 10 • 123k • 434 • 1 PaDT-MLLM/RefCOCO Viewer • Updated Oct 10 • 357k • 463 • 3
PaDT-Paper Patch-as-Decodable-Token: Towards Unified Multi-Modal Vision Tasks in MLLMs Paper • 2510.01954 • Published Oct 2 • 12
Patch-as-Decodable-Token: Towards Unified Multi-Modal Vision Tasks in MLLMs Paper • 2510.01954 • Published Oct 2 • 12
