Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Zero-Shot Compositional Video Learning with Coding Rate Reduction
Heeseok Jung, Jun-Hyeon Bak, Yujin Jeong, Gyugeun Lee, Jinwoo Ahn, Eun-Sol Kim
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 spatiotemporal position embeddings to resampler (ablation) | coding-rate-reduction resampler · Sth-Com (VideoSwin-T) | HM: 28.1 -> 36.8 | Improved |
| add rate reduction optimization step on top of position embeddings (ablation) | coding-rate-reduction resampler · Sth-Com (VideoSwin-T) | HM: 36.8 -> 37.3 | Improved |
| add rate reduction optimization step on top of position embeddings (ablation) | coding-rate-reduction resampler · Sth-Com (VideoSwin-T) | AUC: 18.0 -> 18.3 | Improved |
| add rate reduction optimization step (ablation) | coding-rate-reduction resampler · Sth-Com test split | entanglement coding rate: 508.4 -> 504.3 | Improved |
| use too many latent queries (M=256 vs 128) (ablation) | resampler · Sth-Com (VideoSwin-T) | HM: 37.3 -> 36.6 | Hurt |
| set number of subspaces K to 24 (heads) rather than 6, 12 or 48 (ablation) | resampler · Sth-Com (VideoSwin-T) | HM: 36.5 -> 37.3 | Improved |
| replace label-embedding concatenation and auxiliary training of C2C with the rate reduction resampler | Sth-Com (VideoSwin-T) · Sth-Com (VideoSwin-T + fastText) | HM: 36.7 -> 37.3 vs C2C (Enhanced) | Improved |
| use CLIP+AIM backbone with the proposed resampler instead of C2C | Sth-Com · Sth-Com (CLIP + AIM) | HM: 41.0 -> 42.6 | Improved |
| proposed method versus fine-tuned VLMs on CATER | CATER · CATER unseen mAP | mAP unseen: 37.0 -> 74.0 vs InternVL2-8B | Improved |
| proposed method versus C2C on object accuracy with CLIP | Sth-Com (CLIP + AIM) · Sth-Com (CLIP + AIM) | object accuracy: 56.2 -> 54.7 | Hurt |
Closest papers by meaning
- Prompting Language-Informed Distribution for Compositional Zero-Shot Learning (ECCV 2024 · similarity 0.90)
- Learning Graph Embeddings for Compositional Zero-Shot Learning (CVPR 2021 · similarity 0.89)
- A Dynamic Learning Method towards Realistic Compositional Zero-Shot Learning (AAAI 2024 · similarity 0.89)
- Beyond Image Classification: A Video Benchmark and Dual-Branch Hybrid Discrimination Framework for Compositional Zero-Shot Learning (CVPR 2025 · similarity 0.88)
- Beyond Seen Primitive Concepts and Attribute-Object Compositional Learning (CVPR 2024 · similarity 0.88)
- Retrieval-Augmented Primitive Representations for Compositional Zero-Shot Learning (AAAI 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.
