3DThinkVLA: Endowing Vision-Language-Action Models with Latent 3D Priors via 3D-Thinking-Guided Co-training 文章

ArXiv CS.CV2026-06-04NEWSen作者: Jiaxin Shi, Xidong Zhang, Fucai Zhu, Zhe Li, Siyu Zhu, Weihao Yuan
查看原文 →
关系图谱

3DThinkVLA: Endowing Vision-Language-Action Models with Latent 3D Priors via 3D-Thinking-Guided Co-training · 相关事件

相关事件

3DThinkVLA: Endowing Vision-Language-Action Models with Latent 3D Priors via 3D-Thinking-Guided Co-training
2026-06-04PRODUCT_LAUNCH影响: MEDIUM