Kai Liu

Search for τ− → μ−μ+μ− decays at the LHCb experiment with Run 2 data

Roel Aaij, Ahmed Sameh Wagih Abdelmotteleb, Carlos Abellan Beteta, Fernando Jesus Abudinén et autres

A bstract A search for the lepton-flavour-violating decay τ − → μ − μ + μ − is carried out using data collected by the LHCb experiment between 2016 and 2018 in proton-proton collisions at the LHC at a centre-of-mass energy of …

nl, gb, ch, us, es, se, fr, pl, it, de, lt, br, cn, cz, hu, ua (code pays fourni par la source)

LDA-1B: Scaling Latent Dynamics Action Model via Universal Embodied Data Ingestion

Jiangran Lyu, Kai Liu, Xuheng Zhang, Haoran Liao et autres

Recent robot foundation models largely rely on large-scale behavior cloning, which imitates expert actions but discards transferable dynamics knowledge embedded in heterogeneous embodied data. While the Unified World Model (UWM) formulation has the potential to leverage such diverse data, existing instantiations struggle …

cn (code pays fourni par la source)

Buffer effect of model condensate oil on formation and kinetic inhibition of methane-propane binary hydrates

Yuanyin Wang, Mucong Zi, Chang Guo, Kai Liu et autres

cn (code pays fourni par la source)

Swell correction for sub-bottom profile data using variational mode decomposition

Yibao Xiao, Xishuang Li, Kai Liu, Qingfeng Hua et autres

cn (code pays fourni par la source)

Reliability-Guided Consistency Learning for Robust Semi-Supervised Semantic Segmentation

Qiaoling Xiong, Guozhi Yan, Xiaolu Li, Ke Wang et autres

Semi-Supervised Semantic Segmentation (SSS) has emerged as an effective paradigm to reduce the reliance on dense pixel-wise annotations. However, existing pseudo-labeling approaches often suffer from instability and semantic inconsistency, particularly in complex consumer electronic environments where visual inputs are noisy and heterogeneous. …

cn, hk (code pays fourni par la source)

OMEGA-Avatar: One-shot Modeling of 360° Gaussian Avatars

Zehao Xia, Yiqun Wang, Zhengda Lu, Kai Liu et autres

Creating high-fidelity, animatable 3D avatars from a single image remains a formidable challenge. We identified three desirable attributes of avatar generation: 1) the method should be feed-forward, 2) model a 360° full-head, and 3) should be animation-ready. However, current work addresses only …

LDA-1B: Scaling Latent Dynamics Action Model via Universal Embodied Data Ingestion

Jiangran Lyu, Kai Liu, Xuheng Zhang, Haoran Liao et autres

Recent robot foundation models largely rely on large-scale behavior cloning, which imitates expert actions but discards transferable dynamics knowledge embedded in heterogeneous embodied data. While the Unified World Model (UWM) formulation has the potential to leverage such diverse data, existing instantiations struggle …

OMEGA-Avatar: One-shot Modeling of 360° Gaussian Avatars

Zehao Xia, Yiqun Wang, Zhengda Lu, Kai Liu et autres

Creating high-fidelity, animatable 3D avatars from a single image remains a formidable challenge. We identified three desirable attributes of avatar generation: 1) the method should be feed-forward, 2) model a 360° full-head, and 3) should be animation-ready. However, current work addresses only …