Research map / Continual learning / Class-incremental
Quantized Residuals to Continuous Prompts for Few-Shot Class Incremental Learning in Vision-Language Models
Abhishek Kumar Sinha, Nitant Dube, Soma Biswas
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 DSQ rotation (ablation) | QR-Prompt · CUB200, final session | accuracy (%): 78.46 -> 80.68 | Improved |
| add HPE cross-subspace attention (ablation) | QR-Prompt · CUB200, final session | accuracy (%): 78.42 -> 80.68 | Improved |
| add Prompt Composer (ablation) | QR-Prompt · CUB200, final session | accuracy (%): 80.07 -> 80.68 | Improved |
| use visual-textual residuals instead of visual features (ablation) | QR-Prompt · CUB200, final session | accuracy (%): 78.59 -> 80.68 | Improved |
| use QR-Prompt instead of BiMC | CLIP-based FSCIL · CUB200 | average accuracy (%): 80.49 -> 82.12 | Improved |
| use QR-Prompt instead of BiMC | CLIP-based FSCIL · CIFAR100 | average accuracy (%): 78.45 -> 79.32 | Improved |
| use QR-Prompt instead of VQPrompt | prompt-based FSCIL · CIFAR100 | average accuracy (%): 82.29 -> 79.32 | Hurt |
| use QR-Prompt instead of VQPrompt | prompt-based FSCIL · miniImageNet | performance drop (%): 1.41 -> 1.64 | Hurt |
| increase number of subspaces M from 8 to 32 (K=32) (ablation) | DSQ · CUB200 | average accuracy | Improved |
| increase M beyond 32 (K=32) (ablation) | DSQ · CUB200 | average accuracy | Hurt |
| increase codewords per subspace K (M=32) (ablation) | DSQ · CUB200 | average accuracy | No clear effect |
| use moderate lambda in [0.1, 0.2] versus very small or large (ablation) | DSQ · CUB200 | average accuracy | Mixed |
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 3 of 4 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
- Make Prompts Adaptable: Bayesian Modeling for Vision-Language Prompt Learning with Data-Dependent Prior (AAAI 2024 · similarity 0.88)
- ProAPO: Progressively Automatic Prompt Optimization for Visual Classification (CVPR 2025 · similarity 0.88)
- Pre-trained Vision and Language Transformers Are Few-Shot Incremental Learners (CVPR 2024 · similarity 0.88)
- SEC-Prompt:SEmantic Complementary Prompting for Few-Shot Class-Incremental Learning (CVPR 2025 · similarity 0.88)
- Advancing Cross-domain Discriminability in Continual Learning of Vision-Language Models (NeurIPS 2024 · similarity 0.88)
- Fine-Grained Post-Training Quantization for Large Vision Language Models with Quantization-Aware Integrated Gradients (CVPR 2026 · 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.
