Robotic manipulation has advanced rapidly in mapping visual observations and language instructions to actions. However, reliable manipulation in open-world environments requires robust spatial understanding: the ability to identify targets in complex layouts (spatial reasoning) and adapt to scenes that continuously evo...
Jing-He Wang, Xin-Rui Cao, Duo Wu et al.· 0 citations
EVAN is proposed, a multimodal reinforcement learning framework utilizing EVidence-Ablated Negatives to train LMMs against their own language-prior failures and mitigates modality imbalance and improves over strong RLVR baselines.
Chen-Xi Liu, Tianyi Xiong, Yan-Shuo Chen et al.· 0 citations
This work introduces eVTA, which learns success probabilities from mixed-quality policy rollouts through temporal-difference-style bootstrapping, without expert demonstrations or intermediate annotations, and introduces RL with Evolving Rewards (RLER), a closed-loop framework that adapts eVTA using newly collected roll...
Duo Wu, Hai-Feng Wang, Rongwei Lu et al.· arXiv.org· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.