With the growing deployment of large language models (LLMs) across diverse domains, concerns regarding their safety have grown substantially. LLM unlearning has emerged as a pivotal approach to removing harmful or unlawful content while maintaining utility. Despite increasing interest, the challenges of continual unlea...
Abudukelimu Wuerkaixi, Qi-Zhou Wang, Sen Cui et al.· International Conference on...· 27 citations· ⚡3
Reliable long-horizon prediction remains a central challenge for embodied world models. Although recent generative models can synthesize visually plausible action-conditioned futures, their latent dynamics are typically learned through unconstrained transition modules, which may gradually deviate from coherent physical...
Ke-Zhou Chen, Yu-Wei Zhan, Hao-Ran Liang et al.· 0 citations
Object-centric video models represent scenes with slots, yet exposed geometry can vary in meaning with appearance. In Invariant Slot Attention (ISA), explicit position and scale can disagree with the decoded center and extent; edits can yield unexpected motion or resizing, and replacing appearance can shift geometry. G...
Hao Huang, Zhe-Kai Wang, Xiang Liu et al.· 0 citations
ROCOACH is a world-model-guided coaching framework that uses imagined failures to guide demonstration requests and expert updates, and results show that world models can serve as active coaches, turning imagined failures into targeted supervision for modular policy improvement.
Jia-Jun Liu, Yi-Fan Chen, Yi-Chao Liu et al.· arXiv.org· 1 citation
Object-centric video models represent scenes with slots, yet exposed geometry can vary in meaning with appearance. In Invariant Slot Attention (ISA), explicit position and scale can disagree with the decoded center and extent; edits can yield unexpected motion or resizing, and replacing appearance can shift geometry. G...
Hao Huang, Zhe-Kai Wang, Xiang Liu et al.· 0 citations
Experiments on RoboTwin2.0 show that confidence-guided selection improves post-training efficiency, while dense frame and patch weighting further enhances prediction quality and embodied trajectory consistency compared with scalar reward, progress, and judge-based scoring baselines.
VERDI is proposed, a continual framework for evidence-licensed world model optimization that characterizes each world model through shared inference-time probes to construct an Optimization Fin- gerprint, retrieves relevant prior experience as ranked hypotheses, and validates every candidate under a frozen target-side...
Jun-Yu Wu, Shiqin Nie, Youyi Kou 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.