Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Beyond Alignment: A Generative Matching Paradigm via Flow Matching for Zero-Shot Skeleton-Based Action Recognition
Xuan Liu, Cong Wu, Wei Fang, Zhenhua Feng
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 triplet flow matching loss Ltf (ablation) | ProtoFM (Lfm baseline) · NTU-60 48/12 ZSL | accuracy: 55.1 -> 57.5 | Improved |
| add prototype-guided center loss Lpgc (ablation) | ProtoFM · NTU-60 48/12 ZSL | accuracy: 57.5 -> 69.2 | Improved |
| add PMGR test-time rectification (ablation) | ProtoFM · NTU-60 48/12 ZSL | accuracy: 69.2 -> 73.2 | Improved |
| add PMGR test-time rectification (ablation) | ProtoFM · NTU-120 110/10 ZSL | accuracy: 76.2 -> 82.9 | Improved |
| combine holistic and phased descriptions (ablation) | ProtoFM · NTU-60 55/5 ZSL | accuracy: 82.7 -> 87.3 | Improved |
| combine holistic and phased descriptions vs phased only (ablation) | ProtoFM · NTU-60 48/12 ZSL | accuracy: 65.8 -> 73.2 | Improved |
| increase per-class memory bank capacity M from 20 to 100 (ablation) | ProtoFM PMGR · NTU-60 48/12 / NTU-120 110/10 | accuracy: best at M=80 | Mixed |
| vary ODE inference steps (ablation) | ProtoFM · NTU-60/NTU-120 | accuracy | Mixed |
| ProtoFM vs Neuron on NTU-60 55/5 ZSL | ProtoFM · NTU-60 55/5 ZSL | accuracy: 86.9 -> 87.3 | Improved |
| ProtoFM vs ScoPLe on unseen-class accuracy in GZSL | ProtoFM · NTU-60 48/12 GZSL | unseen accuracy: 61.8 (ScoPLe) vs 54.6 (ProtoFM) | Hurt |
Closest papers by meaning
- Neuron: Learning Context-Aware Evolving Representations for Zero-Shot Skeleton Action Recognition (CVPR 2025 · similarity 0.89)
- Part-aware Unified Representation of Language and Skeleton for Zero-shot Action Recognition (CVPR 2024 · similarity 0.88)
- Boosting Skeleton-based Zero-Shot Action Recognition with Training-Free Test-Time Adaptation (NeurIPS 2025 · similarity 0.88)
- Semantic-guided Cross-Modal Prompt Learning for Skeleton-based Zero-shot Action Recognition (CVPR 2025 · similarity 0.88)
- Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action Recognition (CVPR 2025 · similarity 0.87)
- Heterogeneous Skeleton-Based Action Representation Learning (CVPR 2025 · 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.
