Research map / Learning theory and algorithms / Clustering
Straight-Path Flow Matching for Incomplete Multi-View Clustering
Yiteng Yuan, Junyan Wang, Zheyuan Liu, Hong Jia, Lei Fan, Zhulin Tao, Lianbo Guo
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 |
|---|---|---|---|
| remove cluster-level contrastive alignment loss (ablation) | Straight-path FM · HandWritten tau=0.3 | ACC: 86.50 -> 27.75 | Hurt |
| remove entropy-based alignment loss (ablation) | Straight-path FM · HandWritten tau=0.3 | ACC: 86.50 -> 77.20 | Hurt |
| remove flow prediction loss (ablation) | Straight-path FM · HandWritten tau=0.3 | ACC: 86.50 -> 73.95 | Hurt |
| remove completion consistency loss (ablation) | Straight-path FM · HandWritten tau=0.3 | ACC: 86.50 -> 76.15 | Hurt |
| add flow-matching losses to alignment-only variant (ablation) | Straight-path FM · HandWritten tau=0.3 | ACC: 77.35 -> 86.50 | Improved |
| removing only flow-pred is worse than removing both flow losses (ablation) | Straight-path FM · HandWritten tau=0.3 | ACC: 73.95 (w/o pred) vs 77.35 (w/o both) | Mixed |
| replace diffusion-based DCG with straight-path flow matching | DCG · Five IMVC benchmarks | ACC: Synthetic3D tau=0.1 91.23 -> 94.17; CUB 82.23 -> 86.00 | Improved |
| compare at high missing rate tau=0.5 | Straight-path FM · Fashion | ACC: 90.04 -> 90.09 | No clear effect |
| compare NMI at CUB tau=0.5 | Straight-path FM · CUB | NMI: 72.21 -> 71.74 | Hurt |
| vary number of flow steps (ablation) | Straight-path FM · HandWritten / Synthetic3D | ACC | No clear effect |
| vary lambda1, lambda2 (flow losses weights) (ablation) | Straight-path FM · HandWritten tau=0.3 | ACC | No clear effect |
| vary lambda3, lambda4 (alignment loss weights) (ablation) | Straight-path FM · HandWritten tau=0.3 | ACC | Mixed |
Closest papers by meaning
- Incomplete Multi-view Clustering via Diffusion Contrastive Generation (AAAI 2025 · similarity 0.89)
- Structured Coupling for Flow Matching (NeurIPS 2026 · similarity 0.87)
- Robust Multi-view Clustering against Imperfect Information (NeurIPS 2026 · similarity 0.87)
- Blockwise Flow Matching: Improving Flow Matching Models For Efficient High-Quality Generation (NeurIPS 2025 · similarity 0.87)
- Medusa: A Multi-Scale High-order Contrastive Dual-Diffusion Approach for Multi-View Clustering (CVPR 2025 · similarity 0.87)
- Discrete Flow Matching (NeurIPS 2024 · similarity 0.87)
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.
