Research map / Video / Video generation
Large Motion Model for Unified Multi-Modal Motion Generation
Mingyuan Zhang, Daisheng Jin, Chenyang Gu, Fangzhou Hong, Zhongang Cai, Jingfang Huang, Chongzhi Zhang, Xinying Guo, Lei Yang, Ying He, Ziwei 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 |
|---|---|---|---|
| add random masking in pretraining (ablation) | LMM-Base · HumanML3D | FID: 31.365 -> 0.138 | Improved |
| add random downsampling (frame-rate augmentation) on top of random masking (ablation) | LMM-Base · HumanML3D | MultiModality: 2.214 -> 2.426 | Improved |
| add random downsampling on top of random masking (ablation) | LMM-Base · AMASS-BMLrub 1000 ms | MPJPE (mm): 76.1 -> 73.6 | Improved |
| use ArtAttention instead of SAMI (ablation) | LMM-Base · HumanML3D | FID: 1.866 -> 0.138 | Improved |
| scale model from LMM-Tiny to LMM-Large (ablation) | LMM · HumanML3D | FID: 0.415 -> 0.040 | Improved |
| unified generalist LMM-Tiny versus specialist FineMoGen with similar structure | LMM-Tiny · HumanML3D | FID: 0.151 -> 0.415 | Hurt |
| generalist LMM-Large on short-horizon motion prediction | LMM-Large · AMASS-BMLrub 80 ms | MPJPE (mm): 10.2 (GCNext) vs 12.8 | Hurt |
| generalist LMM-Large on long-horizon motion prediction | LMM-Large · 3DPW 1000 ms | MPJPE (mm): 72.0 -> 68.0 | Improved |
| generalist LMM-Large on music-to-dance quality (FIDk) | LMM-Large · AIST++ | FIDk: 19.01 (TM2D) vs 22.08 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- FineMoGen — FineMoGen: Fine-Grained Spatio-Temporal Motion Generation and Editing (NeurIPS 2023) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Scaling Large Motion Models with Million-Level Human Motions (ICML 2025 · similarity 0.91)
- The Quest for Generalizable Motion Generation: Data, Model, and Evaluation (ICLR 2026 · similarity 0.90)
- AMD: Autoregressive Motion Diffusion (AAAI 2024 · similarity 0.90)
- MotionCraft: Crafting Whole-Body Motion with Plug-and-Play Multimodal Controls (AAAI 2025 · similarity 0.88)
- Human Motion Diffusion as a Generative Prior (ICLR 2024 · similarity 0.88)
- MotionGPT: Finetuned LLMs Are General-Purpose Motion Generators (AAAI 2024 · similarity 0.88)
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.
