Research map / Video / Video generation
Training-free Controllable Motion Generation under Heterogeneous Constraints
Xiaofei Hui, Bo Yan, Haoxuan Qu, Hossein Rahmani, Jun Liu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use forward-evaluation control for criterion constraints instead of surrogate gradients (ablation) | MIC · HSI-2 | unsuccess rate: 0.219 -> 0.094 | Improved |
| use unified handling of both constraint types vs separate (ablation) | MIC · HSI-2 | sim pass rate: 0.656 -> 0.875 | Improved |
| add feedback regulator (ablation) | MIC · HSI-2 | sim pass rate: 0.813 -> 0.875 | Improved |
| add control allocator (ablation) | MIC · HSI-2 | constraint error: 0.048 -> 0.009 | Improved |
| add full coordination mechanism (ablation) | MIC · HSI-2 | sim pass rate: 0.688 -> 0.875 | Improved |
| MIC vs best gradient baselines | MIC · GEO-1 (open-set) | unsuccess rate: 0.531 -> 0.094 | Improved |
| MIC vs baselines on motion quality | MIC · HSI-1 (known) | FID: 0.556 -> 0.494 | Improved |
| MIC vs IK on skating | MIC · HSI-2 (open-set) | foot skating: 0.253 -> 0.172 | Improved |
Closest papers by meaning
- Programmable Motion Generation for Open-Set Motion Control Tasks (CVPR 2024 · similarity 0.86)
- Generative Control as Optimization: Time Unconditional Flow Matching for Adaptive and Robust Robotic Control (NeurIPS 2026 · similarity 0.86)
- OptiWorld: Optimal Control for Video World Generation under Physical Constraints (NeurIPS 2026 · similarity 0.85)
- Absolute State-wise Constrained Policy Optimization: High-Probability State-wise Constraints Satisfaction (NeurIPS 2026 · similarity 0.85)
- Understanding and Improving Training-free Loss-based Diffusion Guidance (NeurIPS 2024 · similarity 0.84)
- One-Step Diffusion Policy: Fast Visuomotor Policies via Diffusion Distillation (ICML 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
