Retrieving and Refining Winning Noise Tickets for Diffusion-Based Motion Generation
Quick Answer
The WINRO framework enhances text-to-motion alignment in diffusion-based models like MDM and MotionLCM by optimizing 'winning noise tickets' for improved semantic consistency.
Quick Take
The WINRO framework enhances text-to-motion alignment in diffusion-based models like MDM and MotionLCM by optimizing 'winning noise tickets' for improved semantic consistency. This approach, which does not require retraining, shows significant fidelity improvements on HumanML3D and robustness on the MTT benchmark, enabling applications in motion stylization and spatial constraints.
Key Points
- WINRO is a training-free, model-agnostic framework for motion generation.
- It retrieves and refines noise tickets to improve text-motion fidelity.
- Demonstrated improvements on HumanML3D and MTT benchmarks.
- Supports applications like motion stylization and spatial constraints.
- Achieves better temporal robustness without retraining existing models.
Paper Resources
📖 Reader Mode
~2 min readAbstract:Diffusion-based text-to-motion models synthesize realistic human motions but often exhibit semantic drift from the input text. Motion is inherently temporal, especially in compositional and long-duration sequences that require semantic consistency across multiple action segments and smooth kinematic transitions throughout the trajectory. We posit that the initial noise is central to this consistency: within the Gaussian noise space, certain instances, i.e. winning noise tickets, carry latent structure that biases denoising toward particular motion semantics, even under null prompts. We propose WInning Noise Retrieval and Optimization (WINRO), a training-free, model-agnostic framework that improves text-motion alignment by selecting and refining such tickets before diffusion sampling. WINRO maps random noises to motion features generated under null prompts, retrieves the best-aligned noise for a given text, and refines it via a KL-regularized objective that reduces the residual semantic gap while preserving the Gaussian prior. An optional LoRA-based adapter amortizes this refinement into a single forward pass. WINRO consistently improves text-motion fidelity across different base models, MDM and MotionLCM, on HumanML3D without retraining, improves temporal robustness on the MTT benchmark, and generalizes to applications such as motion stylization and spatial constraint satisfaction.
| Comments: | Accepted to ECCV 2026, Project page: this https URL |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2607.06843 [cs.CV] |
| (or arXiv:2607.06843v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2607.06843 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Sakuya Ota [view email]
[v1]
Tue, 7 Jul 2026 22:32:10 UTC (10,591 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →ProMoE-FL: Prototype-conditioned Mixture of Experts for Multimodal Federated Learning with Missing Modalities
ProMoE-FL introduces a Prototype-conditioned Mixture-of-Experts framework for multimodal federated learning, effectively addressing missing modalities. It outperforms existing methods on four chest X-ray datasets, demonstrating superior feature synthesis capabilities in both homogeneous and heterogeneous settings.