Research map / Continual learning / Continual fine-tuning of LLMs
A Unified Continual Learning Framework with General Parameter-Efficient Tuning
Qiankun Gao, Chen Zhao, Yifan Sun, Teng Xi, Gang Zhang, Bernard Ghanem, Jian 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 learning with calibrated adaptation speed (ablation) | LAE Adapter10 · ImageNet-R | A10: 65.22 -> 72.45 | Improved |
| add accumulation and ensemble on top of calibrated learning (ablation) | LAE Adapter10 · ImageNet-R | A10: 70.30 -> 72.45 | Improved |
| use expert ensemble instead of offline PET only (ablation) | LAE Adapter10 · ImageNet-R | A10: 72.80 -> 72.45 (A10); 78.63 -> 79.07 (average) | No clear effect |
| add EMA accumulation to the ensemble (ablation) | LAE Adapter10 without calibrated learning · ImageNet-R | A10: 63.36 -> 65.22 | Improved |
| add gradient compensation to Prefix (ablation) | LAE Prefix20 · ImageNet-R | A10: 69.23 -> 70.53 | Improved |
| add learnable scaling parameters to Prefix (ablation) | LAE Prefix20 · ImageNet-R | A10: 69.23 -> 70.38 | Improved |
| combine compensation and scale for Prefix (ablation) | LAE Prefix20 · ImageNet-R | A10: 69.23 -> 72.05 | Improved |
| apply LAE framework to baseline PET with Adapter5 | Adapter5 PET · ImageNet-R | A10: 61.63 -> 72.66 | Improved |
| apply LAE to baseline with Prefix10 | Prefix10 PET · CIFAR100 | A10: 84.49 -> 85.17 | Improved |
| increase PET module size (ablation) | LAE · ImageNet-R | A10: 72.66 -> 72.45 (Adapter5 -> Adapter10) | No clear effect |
| use LAE with Swin-B instead of baseline | Swin-B Adapter · ImageNet-R | A10: 71.81 -> 73.38 | Improved |
| use LAE with ConvNeXt-B instead of baseline | ConvNeXt-B Adapter · ImageNet-R | A10: 76.35 -> 78.38 | 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.
- DualPrompt — DualPrompt: Complementary Prompting for Rehearsal-Free Continual Learning (ECCV 2022) · better on 2 of 2 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
- Consistent Prompting for Rehearsal-Free Continual Learning (CVPR 2024 · similarity 0.88)
- Training Consistent Mixture-of-Experts-Based Prompt Generator for Continual Learning (AAAI 2025 · similarity 0.88)
- Evolving Parameterized Prompt Memory for Continual Learning (AAAI 2024 · similarity 0.88)
- Mixture of Experts Meets Prompt-Based Continual Learning (NeurIPS 2024 · similarity 0.88)
- REP: Resource-Efficient Prompting for Rehearsal-Free Continual Learning (NeurIPS 2025 · similarity 0.88)
- SAFE: Slow and Fast Parameter-Efficient Tuning for Continual Learning with Pre-Trained Models (NeurIPS 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.
