Research map / Video / Action recognition
Recurring the Transformer for Video Action Recognition
Jiewen Yang, Xingbo Dong, Liujun Liu, Chao Zhang, Jiajun Shen, Dahai Yu
This paper tried a common change: Add residual connections
Add a hidden-state residual connection · Jester-V1 · Accuracy (%): 88.7 -> 91.3 Improved
What 546 accepted papers report for this change
- 449 improved
- 35 no clear effect
- 11 mixed
- 51 hurt
- Resolution as a Direction: Vector-Panning Feature Alignment for Cross-Resolution Re-Identification ICML 2026 · add residual connection to VP · MLR-Market-1501 · Rank-1 (%): 91.2 -> 94.1Improved
- MAGIC: Multi-Granularity Language-Informed Image Clustering ICML 2026 · add residual connection to semantic adapters · CIFAR-20 · ACC: 66.1 -> 59.9Hurt
- Domain Adaptive Object Detection via Dynamic Causal Refinement ICML 2026 · add residual connection to SCC · Cityscapes to Foggy Cityscapes (0.02 split) · mAP@50: 64.0 -> 64.6Improved
See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.
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 |
|---|---|---|---|
| Use recurrent frame-by-frame processing with an attention gate that carries temporal features in a hidden state | RViT · Kinetics-400 | Top-1 accuracy (%): 81.3 -> 81.5 | Improved |
| Use recurrent frame-by-frame processing with an attention gate that carries temporal features in a hidden state | RViT · Something-Something-V2 | Top-1 accuracy (%): 67.7 -> 67.9 | Improved |
| Use recurrent frame-by-frame processing with an attention gate that carries temporal features in a hidden state | RViT · Jester | Top-1 accuracy (%): 90.75 -> 92.31 | Improved |
| Use recurrent frame-by-frame processing with an attention gate that carries temporal features in a hidden state | RViT · Charades | mAP (%): 63.2 -> 66.1 | Improved |
| Add both spatial Ts and temporal Tt tokens for classification (ablation) | RViT · Jester val | Accuracy (%): 79.94 -> 92.31 | Improved |
| Add both spatial Ts and temporal Tt tokens for classification (ablation) | RViT · Jester val | Accuracy (%): 88.16 -> 92.31 | Improved |
| Use linear attention instead of softmax attention (ablation) | RViT◦ with residual connection · Jester-V1 | Accuracy (%): 87.9 -> 91.3 | Improved |
| Add a hidden-state residual connection (ablation) | RViT◦ with linear attention · Jester-V1 | Accuracy (%): 88.7 -> 91.3 | Improved |
Closest papers by meaning
- Recurrent Video Masked Autoencoders (CVPR 2026 · similarity 0.89)
- Recurrent Vision Transformers for Object Detection With Event Cameras (CVPR 2023 · similarity 0.88)
- Adapting Short-Term Transformers for Action Detection in Untrimmed Videos (CVPR 2024 · similarity 0.86)
- Recurrent Video Restoration Transformer with Guided Deformable Attention (NeurIPS 2022 · similarity 0.86)
- Self-Supervised Video Transformer (CVPR 2022 · similarity 0.86)
- Streaming Video Model (CVPR 2023 · 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.
