UniMate: One Unified Model to Animate Diverse Skeletons

Linzhan Mou, Jiahui Lei, Zhiyang Dou, Chenyue Cai, Chaoyue Song et al.|September 4, 2026arXiv

Key Takeaway

For the first time, a single model can animate any skeleton topology (bipedal, quadrupedal, insects, etc.) from text alone—no per-character fine-tuning or reference motions needed at inference time.

Summary

UniMate is a foundation model that generates realistic motion for any 3D character skeleton from text descriptions, without needing to retrain for each new character type. It uses a specialized neural architecture that understands skeleton structure through graph-based attention mechanisms, and was trained on a diverse dataset of 13,000+ motion sequences across different creature types.

multimodal architecture

Key Terms

diffusion-transformer graph-laplacian rotary-positional-encoding topology-aware-attention zero-shot-cross-topology-transfer