Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Task Discrepancy Maximization for Fine-Grained Few-Shot Classification
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 TDM | ProtoNet and other metric-based few-shot classifiers · Fine-grained few-shot benchmarks | accuracy: 62.90 -> 69.94 | Helped |
| Add TDM | FRN · tiered meta-iNat, 5-shot | accuracy: 63.45 -> 62.91 | Hurt |
| Add SAM to ProtoNet (ablation) | ProtoNet · CUB_cropped, 1-shot | accuracy: 62.90 -> 68.53 | Helped |
| Add SAM to ProtoNet (ablation) | ProtoNet · Aircraft, 5-shot | accuracy: 68.96 -> 69.33 | Helped |
| Add QAM to ProtoNet (ablation) | ProtoNet · CUB_cropped, 1-shot | accuracy: 62.90 -> 65.11 | Helped |
| Add QAM to ProtoNet (ablation) | ProtoNet · Aircraft, 5-shot | accuracy: 68.96 -> 70.85 | Helped |
| Combine SAM and QAM rather than using either alone (ablation) | ProtoNet · CUB_cropped and Aircraft | accuracy: CUB_cropped 1-shot: 68.53 -> 69.94; Aircraft 5-shot: 70.85 -> 71.12 | Helped |
| Use average pooling instead of max pooling to predict discriminative channels (ablation) | TDM · Fine-grained few-shot classification | accuracy | Helped |
Closest papers by meaning
- CAD: Co-Adapting Discriminative Features for Improved Few-Shot Classification (CVPR 2022 · similarity 0.88)
- Learning a Discriminative Filter Bank Within a CNN for Fine-Grained Recognition (CVPR 2018 · similarity 0.87)
- Discriminative Sample-Guided and Parameter-Efficient Feature Space Adaptation for Cross-Domain Few-Shot Learning (CVPR 2024 · similarity 0.85)
- Pushing the Limit of Fine-Tuning for Few-Shot Learning: Where Feature Reusing Meets Cross-Scale Attention (AAAI 2024 · similarity 0.85)
- Few-Shot Recognition via Stage-Wise Retrieval-Augmented Finetuning (CVPR 2025 · similarity 0.85)
- Unlabeled Data Improves Fine-Grained Image Zero-shot Classification with Multimodal LLMs (NeurIPS 2025 · similarity 0.84)
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.