Research map / Continual learning / Online and replay-based
Decomposing Prompts, Composing Actions: A Multi-Granularity Prompting Approach for Incremental Action Learning
Xinyi Cheng, Chenghao Xu, Xi Wang, Jiexi Yan, Yanhua Yang
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 temporal prompt pool Pt (ablation) | DPCA · NTU RGB+D CS, final session | AHM: 46.8 -> 57.7 | Improved |
| add spatial prompt pool Ps (ablation) | DPCA · NTU RGB+D CS, final session | AHM: 52.0 -> 57.7 | Improved |
| use multi-granularity prompts instead of single comprehensive prompt (ablation) | DPCA · NTU RGB+D CS, final session | AHM: 40.6 -> 57.7 | Improved |
| add task-agnostic prompt pool Pa (ablation) | DPCA · NTU RGB+D CS, final session | AHM: 54.1 -> 57.7 | Improved |
| add Dissimilarity Attention Rectification (ablation) | DPCA · NTU RGB+D CS, final session | AHM: 56.7 -> 57.7 | Improved |
| add 3 prompts per task instead of 2 (ablation) | DPCA · NTU RGB+D CS, final session | AHM: 57.0 -> 57.7 | Improved |
| add 5 prompts per task instead of 3 (ablation) | DPCA · NTU RGB+D CS, final session | AHM: 57.7 -> 56.2 | Hurt |
| use DPCA instead of POET on PKU-MMD | POET · PKU-MMD | AHM: 55.2 -> 56.5 | Improved |
| use DPCA instead of POET on SHREC-2017 | POET · SHREC-2017 | AHM: 55.8 -> 57.3 | Improved |
| use DPCA instead of POET with ST-GCN backbone | POET · NTU RGB+D CS (ST-GCN) | AHM: 54.3 -> 55.7 | Improved |
| use DPCA instead of POET for new-class learning | POET · NTU RGB+D CS, final session | New-class accuracy: 55.8 -> 55.3 | No clear effect |
| use image-style prompt methods (L2P, CODA-P) for skeleton continual learning | frozen CTR-GCN · NTU RGB+D CS, final session | AHM: 57.7 -> 0.0 (L2P) | Hurt |
Closest papers by meaning
- CODA-Prompt: COntinual Decomposed Attention-Based Prompting for Rehearsal-Free Continual Learning (CVPR 2023 · similarity 0.87)
- CE$^4$L: Continual Ego, Exo, and Ego-Exo Learning (ICML 2026 · similarity 0.87)
- Convolutional Prompting meets Language Models for Continual Learning (CVPR 2024 · similarity 0.86)
- Mixture of Experts Meets Prompt-Based Continual Learning (NeurIPS 2024 · similarity 0.86)
- Evolving Parameterized Prompt Memory for Continual Learning (AAAI 2024 · similarity 0.86)
- Advancing Prompt-Based Methods for Replay-Independent General Continual Learning (ICLR 2025 · similarity 0.85)
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.
