Research map / Video / Video generation
Human Motion Unlearning
Edoardo De Matteis, Matteo Migliarini, Alessio Sampieri, Indro Spinelli, Fabio Galasso
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 |
|---|---|---|---|
| apply Latent Code Replacement (LCR) | MoMask · HumanML3D retain set (safe) | FID: 0.075 -> 0.050 vs Dr; original 0.041 | Improved |
| apply LCR instead of UCE | MoMask · HumanML3D retain set (safe) | FID: 0.090 -> 0.050 | Improved |
| apply ESD (fine-tuning) unlearning | MoMask · HumanML3D retain set (safe) | FID: 0.041 -> 30.679 | Hurt |
| apply UCE unlearning | BAMM · HumanML3D retain set (safe) | FID: 0.181 -> 0.257 | Hurt |
| apply LCR | BAMM · HumanML3D forget set | MM-Safe: 4.995 -> 4.647 | Improved |
| apply LCR | MoMask · Motion-X retain set (safe) | FID: 3.697 -> 3.658 | No clear effect |
| apply UCE unlearning | MoMask · Motion-X forget set (violent) | MM-Safe: 9.291 -> 14.470 | Hurt |
| apply ESD unlearning | MoMask · Motion-X retain set (safe) | FID: 3.697 -> 172.55 | Hurt |
| unlearn only kick codes with LCR (ablation) | MoMask · Motion-X forget set (violent) | FID: 8.331 -> 36.995 | Improved |
| unlearn only stab with LCR (ablation) | MoMask · Motion-X forget set (violent) | FID: 8.331 -> 24.315 | No clear effect |
| use LCR with implicit (rephrased) prompts | MoMask · HumanML3D implicit kick prompts | violence suppression | Improved |
| apply RECE instead of UCE | BAMM · HumanML3D forget set | MM-Safe: 6.581 -> 4.965 | Improved |
Closest papers by meaning
- Breaking The Limits of Text-conditioned 3D Motion Synthesis with Elaborative Descriptions (ICCV 2023 · similarity 0.87)
- MotionMix: Weakly-Supervised Diffusion for Controllable Motion Generation (AAAI 2024 · similarity 0.87)
- AMD: Autoregressive Motion Diffusion (AAAI 2024 · similarity 0.87)
- Scaling Large Motion Models with Million-Level Human Motions (ICML 2025 · similarity 0.86)
- MMM: Generative Masked Motion Model (CVPR 2024 · similarity 0.86)
- Seamless Human Motion Composition with Blended Positional Encodings (CVPR 2024 · similarity 0.86)
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.
