We study post-training for dynamically routed mixture-of-experts (MoE) models, focusing on integrating independently optimized specialists while keeping consolidation efficient. Our approach combines calibrated multi-teacher on-policy distillation (C-MOPD), which matches each selected teacher on a local Top-$K$ support...
Yue He, Jia-Hui Hu, Sha-Man Tang et al.· 0 citations
Vision-language robot manipulation policies can follow semantic instructions, but adapting them to a new procedure from only a few demonstrations remains difficult because language underspecifies contact timing, motion phases, corrective behavior, and execution style. This paper presents Task-Prototype Guided Flow Matc...
Open-vocabulary referring segmentation in 3D Gaussian Splatting (3DGS) requires a neural model to select Gaussian primitives according to free-form language expressions. Existing 3DGS-based methods usually rely on global text-region similarity, which is weak for queries involving attributes, reference objects, spatial...
Yi-Zhao Wang, Xin-Fa Wang, Jing-Bo Wang et al.· 0 citations
A structure-aware rectified flow model with patch-position conditioning, global-local coupled velocity prediction, and density-aware velocity weighting is designed, enabling direct latent generation of class-conditioned 3DGS objects within seconds.
Despite its compact active-parameter budget, Turing-20B-A2B achieves, at the base-model stage, overall general capability exceeding Qwen3-8B Base and approaching Qwen3.5-9B Base, while maintaining strong long-context performance and favorable prefill-latency scaling.
Yu-Heng Zhang, Yi-Zhao Wang, Daijun Zhu 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.