Style-Conditioned Motion Generation

Follow this topic Ask about this topic

Latest papers

10 papers3 papers in the last four weeks, level with the four weeks before. 0.0% of all new papers.

Past as State, Present as Attention: Persistent-State Blockwise Flow Matching for Long-Horizon Co-Speech Motion Generation
Oct 23, 2026Yi Liu, Xiangyue Zhang, Jianfang Li +3Human Motion Generation Co-Speech Gesture Generation
Tongji University · Alibaba Group · The University of Tokyo
Walk Like This: Stylized Locomotion for Physics-Based Characters through Example-Driven Diffusion
Oct 13, 2026Joshua Aurand, Agon Serifi, Sammy Christen +4Style-Conditioned Motion Generation Robot Locomotion
Disney Research Zürich, Switzerland · ETH Zürich, Switzerland
VICAR: Video Imitation with Contact-Aware Retargeting for Humanoid Interactive Skills
Oct 9, 2026Dvij Kalaria, Bike Zhang, S. Shankar SastryStyle-Conditioned Motion Generation Robot Imitation Learning
University of California, Berkeley
FlowHMR: Physically Plausible Motion Capture from Video
Oct 2, 2026Zhanke Wang, Chengfeng Zhao, Qing Shuai +8Style-Conditioned Motion Generation 3D Human Motion Reconstruction
Peking University · The Hong Kong University of Science and Technology · Tencent +3
GestAdapt: Workspace-Conditioned Co-Speech Gesture Generation for Humanoid Robots
Sep 29, 2026Bosong Ding, Xianglin Zhang, Miao Xin +2Robot Motion Generation Style-Conditioned Motion Generation
Department of Intelligent Systems, Tilburg University, The Netherlands. · TU-Dresden, Dresden, Germany. · School of Artificial Intelligence, China University of Mining and Technology-Beijing (CUMTB), Beijing, China.
Flexible Motion Generation from Language and Style References
Sep 7, 2026Kai Weixian Lan, Bodie Criswell, Briana Fedkiw +3Text-to-Motion Generation Motion Diffusion Models
University of California, Davis, USA · University of Utah, USA · Epic Games, USA +1
Prior-First, Condition-Second: Scalable and Controllable Hand Motion Completion
Jul 7, 2026Mingyi Shi, Xuelin Chen, Taku KomuraAutoregressive Motion Generation Human Motion Generation
The University of Hong Kong, Hong Kong · Adobe Research, London, UK
HO-Flow: Generalizable Hand-Object Interaction Generation with Latent Flow Matching
Apr 12, 2026Zerui Chen, Rolandos Alexandros Potamias, Shizhe Chen +3Style-Conditioned Motion Generation HOI Generation
Imperial College London · Inria, CNRS, ENS-PSL
Sketch2Colab: Sketch-Conditioned Multi-Human Animation via Controllable Flow Distillation
Mar 2, 2026Divyanshu Daiya, Aniket BeraMarkov Chains Style-Conditioned Motion Generation
IDEAS Lab, Department of Computer Science, Purdue University
Robo-Saber: Generating and Simulating Virtual Reality Players
Feb 20, 2026Nam Hee Kim, Jingjing May Liu, Jaakko Lehtinen +3Human Behavior Simulation Human Motion Generation
Aalto University, Finland · University of California, Berkeley, United States · NVIDIA +1

Oct 23, 2026J/K move · Enter open · S save

Past as State, Present as Attention: Persistent-State Blockwise Flow Matching for Long-Horizon Co-Speech Motion Generation

Yi Liu, Xiangyue Zhang, Jianfang Li, Jia Ma+2

Tongji University · Alibaba Group · The University of Tokyo

Human Motion Generation Co-Speech Gesture Generation Style-Conditioned Motion Generation

Open paper AskPDF arXiv

Co-speech motion generation remains challenging over long horizons due to the need for temporal coherence and consistent style. Existing methods typically generate long motions segment by segment, either conditioning on a short seed of recent frames, which gradually drifts, or retaining ever-longer explicit histories, which are costly to maintain and hard to learn from limited data. We trace these limitations to a key asymmetry between past and present: past motion carries slowly varying style that should be compactly summarized, whereas the current segment exhibits fast-changing local dynamics that require fine-grained modeling. This asymmetry motivates a simple Past-As-State, Present-as-Attention principle for long-horizon co-speech motion generation. Following this principle, we propose PASPA, which summarizes past history with an inter-block persistent state and models the current segment with intra-block bidirectional self-attention. We instantiate PASPA within a blockwise flow matching framework, with multi-block supervision for effective training over variable histories, prefill-decode state caching for efficient rollout, and hybrid classifier-free guidance for enhanced history and speech conditioning. Extensive public-benchmark experiments show that PASPA achieves state-of-the-art FGD performance and preserves motion style across long-sequence generation.