Research map / Video / Action recognition
Learning to Generalize without Bias for Open-Vocabulary Action Recognition
Yating Yu, Congqi Cao, Yifan Zhang, Yanning Zhang
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 cross-batch meta-optimization (ablation) | VCLIP · UCF-SCUBA top-1 | accuracy: 28.3 -> 32.1 | Improved |
| add Gaussian Weight Average self-ensemble (ablation) | VCLIP · UCF-SCUBA top-1 | accuracy: 28.3 -> 30.7 | Improved |
| combine meta-optimization and GWA (ablation) | VCLIP · UCF-SCUBA top-1 | accuracy: 28.3 -> 33.5 | Improved |
| use cross-batch meta-optimization instead of gradient accumulation (ablation) | VCLIP · UCF-SCUBA top-1 | accuracy: 28.9 -> 33.5 | Improved |
| add residual feature distillation CLIP regularization (ablation) | VCLIP · UCF-SCUBA top-1 | accuracy: 28.3 -> 25.2 | Hurt |
| add interpolated weight regularization (ablation) | VCLIP · UCF-SCUBA top-1 | accuracy: 28.3 -> 28.9 | No clear effect |
| use cross-batch meta-optimization instead of MAML meta unseen (ablation) | VCLIP · UCF-SCUBA top-1 | accuracy: 31.8 -> 33.5 | Improved |
| shuffle batch sampler instead of initial order (ablation) | Open-MeDe · UCF-SCUBA top-1 | accuracy: 28.9 -> 33.5 | Improved |
| use shuffled sampler instead of semantically similar classes across batches (ablation) | Open-MeDe · UCF-SCUBA top-1 | accuracy: 30.9 -> 33.5 | Improved |
| add CLIP prediction-space ensemble | Open-MeDe · UCF-SCUBA top-1 | accuracy: 33.5 -> 32.0 | Hurt |
| add CLIP weight-space ensemble | Open-MeDe · UCF-SCUBA top-1 | accuracy: 33.5 -> 29.9 | Hurt |
| integrate Open-MeDe into X-CLIP partially tuned learner | X-CLIP · UCF cross-dataset | top-1 accuracy: 72.0 -> 79.3 | Improved |
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.
- FROSTER — FROSTER: Frozen CLIP is A Strong Teacher for Open-Vocabulary Action Recognition (ICLR 2024) · better on 9 of 9 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
- FROSTER: Frozen CLIP is A Strong Teacher for Open-Vocabulary Action Recognition (ICLR 2024 · similarity 0.91)
- TACO: Towards Task-Consistent Open-Vocabulary Adaptation in Video Recognition (NeurIPS 2026 · similarity 0.88)
- Open-Vocabulary Video Relation Extraction (AAAI 2024 · similarity 0.86)
- Does Video-Text Pretraining Help Open-Vocabulary Online Action Detection? (NeurIPS 2024 · similarity 0.86)
- OST: Refining Text Knowledge with Optimal Spatio-Temporal Descriptor for General Video Recognition (CVPR 2024 · similarity 0.86)
- Are We Making Progress in Multimodal Domain Generalization? A Comprehensive Benchmark Study (NeurIPS 2026 · 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.
