Ant Lingbo Unveils LingBot-VA 2.0 World Action Model on July 10

According to Pandaily, Ant Lingbo unveiled LingBot-VA 2.0, an embodied-native world action model, on July 10. Unlike industry-standard approaches using video generation model fine-tuning, LingBot-VA 2.0 employs autoregressive architecture trained from scratch, designed specifically for dynamic modeling, causal prediction, and real-time execution in robot-environment interactions. Ant Lingbo Chief Scientist Shen Yujun stated that the field of embodied AI remains immature with many unresolved challenges. CEO Zhu Xing noted that model deployment faces significant difficulties in both pretraining and post-training phases, indicating the industry is still in early commercialization stages.
Disclaimer: The information on this page may come from third-party sources and is for reference only. It does not represent the views or opinions of Gate and does not constitute any financial, investment, or legal advice. Virtual asset trading involves high risk. Please do not rely solely on the information on this page when making decisions. For details, see the Disclaimer.
Comment
0/400
No comments