World-action models (WAMs) offer a promising framework for embodied intelligence, yet mainstream approaches remain centered on video foundation models or vision-language models. Even when geometric features extracted by 3D models such as VGGT are incorporated, they are typically passed into a video model or VLM that se...
Xiao Lai, Zhi-Hao Zhan, Zhi-Quan Liao et al.· 1 citation
A semi-supervised deep image-stitching method for moving elongated objects that combines semi-supervised registration and unsupervised reconstruction, and designs a reconstruction objective function based on multi-scale feature representations.
Xiao Lai, Zi-Qi Xie, Xianhui Liu· Italian National Conference...· 0 citations
This paper proposes Collaborative Weighting Actor-Critic (CWAC), a unified framework that explicitly accounts for predictive uncertainty in value estimation and introduces a collaborative weighting mechanism that jointly reweights TD-errors and uncertainty, enabling robust learning from reliable samples while suppressi...
Gong Gao, Xiao Lai, Zi-Qi Xie et al.· arXiv.org· 0 citations
This work shows that enforcing high-quality behavioral support during online learning effectively mitigates critic-induced error amplification, enabling structured behavior priors to guide policy updates in a principled and sample-efficient manner.
Gong Gao, Xiao Lai, Jia-Ji Shen et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.