The spatial intelligence of large vision-language models (LVLMs) requires the ability to combine local visual observations into a geometrically consistent representation of the physical world. Many static benchmarks assess isolated spatial judgments, such as relative direction or distance, while interactive embodied ev...
Jun-Fei Wu, Qiu-Yue Wang, Xu-Hong Huang et al.· 0 citations
Reinforcement learning for large language model post-training often reuses each batch of sampled rollouts for several gradient passes to amortize the cost of generation. Under such multi-pass reuse, existing objectives fail in different ways: unclipped updates collapse as the policy drifts away from the sampling policy...
Xunlan Zhou, Yu-Zheng Zhao, Xu-Hong Huang et al.· 0 citations
FineVLA, an open framework for action-aligned fine-grained VLA supervision, is introduced and the largest real-world gains appear on pose, color, and approach direction, and approach direction--factors where goal-level instructions provide no guidance.