Research map / Image generation / Fast diffusion sampling
Few-step Flow for 3D Generation via Marginal-Data Transport Distillation
Zanwei Zhou, Taoran Yi, Jiemin Fang, Chen Yang, Lingxi Xie, Xinggang Wang, Wei Shen, Qi Tian
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 velocity matching loss to 1-step TRELLIS distillation (ablation) | MDT-dist · Toys4k | FD_dinov2: 195.6 -> 172.0 | Improved |
| add velocity distillation on top of velocity matching (ablation) | MDT-dist · Toys4k | FD_dinov2: 172.0 -> 164.2 | Improved |
| add velocity distillation on top of velocity matching (ablation) | MDT-dist · Toys4k | ULIP-I: 35.99 -> 36.88 | Improved |
| use MDT-dist instead of consistency distillation (sCM) | TRELLIS 1-step · Toys4k | FD_dinov2: 186.0 -> 164.2 | Improved |
| increase student steps from 1x2 to 2x2 | MDT-dist · Toys4k | FD_dinov2: 164.2 -> 110.9 | Improved |
| distill 25x2-step TRELLIS to 1-2 steps | TRELLIS · Toys4k | FD_dinov2: 65.24 -> 110.9 | Hurt |
Closest papers by meaning
- Consistent Flow Distillation for Text-to-3D Generation (ICLR 2025 · similarity 0.89)
- Shortcutting Pre-trained Flow Matching Diffusion Models is Almost Free Lunch (NeurIPS 2025 · similarity 0.89)
- D-Flow: Differentiating through Flows for Controlled Generation (ICML 2024 · similarity 0.88)
- Diff2Flow: Training Flow Matching Models via Diffusion Model Alignment (CVPR 2025 · similarity 0.88)
- Consistent3D: Towards Consistent High-Fidelity Text-to-3D Generation with Deterministic Sampling Prior (CVPR 2024 · similarity 0.88)
- Diffusion Time-step Curriculum for One Image to 3D Generation (CVPR 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.
