Model checkpoints for UniAR: Unified Multimodal Autoregressive Modeling with Shared Context—Visual Tokenizer is Key to Unification.
AI & ML interests
ShareLab-SII is a research group at the Shanghai Innovation Institute (SII) led by Prof. Zuxuan Wu. SII is an institution dedicated to innovation in education and research in the field of AI. We work on computer vision and deep learning, with interests in: large-scale visual understanding and generation efficient visual architectures embodied intelligence
Recent Activity
Datasets for ThinkingVLA: Interleaved Vision and Language Reasoning for Robotic Manipulation.
-
ShareLab-SII/thinking_jaco_play
Viewer • Updated • 69.7k • 525 -
ShareLab-SII/thinking_berkeley_cable_routing_lerobot_gemini_output
Viewer • Updated • 38.2k • 760 -
ShareLab-SII/thinking_berkeley_cable_routing_lerobot_output_qwen3vl
Viewer • Updated • 38.2k • 933 -
ShareLab-SII/thinking_utaustin_mutex_lerobot_output_qwen3vl
Viewer • Updated • 362k • 782
Model checkpoints for UniAR: Unified Multimodal Autoregressive Modeling with Shared Context—Visual Tokenizer is Key to Unification.
Datasets for ThinkingVLA: Interleaved Vision and Language Reasoning for Robotic Manipulation.
-
ShareLab-SII/thinking_jaco_play
Viewer • Updated • 69.7k • 525 -
ShareLab-SII/thinking_berkeley_cable_routing_lerobot_gemini_output
Viewer • Updated • 38.2k • 760 -
ShareLab-SII/thinking_berkeley_cable_routing_lerobot_output_qwen3vl
Viewer • Updated • 38.2k • 933 -
ShareLab-SII/thinking_utaustin_mutex_lerobot_output_qwen3vl
Viewer • Updated • 362k • 782