Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Task-Adaptive Negative Envision for Few-Shot Open-Set Recognition
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 |
|---|---|---|---|
| Use task-adaptive negative prototypes instead of manually tuned rejection thresholds | FSOR classifier · CIFAR-FS and FC100 | AUROC | Helped |
| Add class semantics to the ATT-G negative generator (ablation) | ATT-G · CIFAR-FS and FC100 | AUROC: 76.72 -> 78.10; 59.55 -> 59.73 | Helped |
| Use multiple negative prototypes (M=5) rather than one (M=1) (ablation) | Negative prototype generator · FSOR | macro-averaged F1-score | Helped |
| Use conjugate training with mutual supervision between paired tasks (ablation) | FSOR and GFSOR methods · FSOR benchmarks | all reported metrics | Helped |
| Use an attention-based negative generator to model inter-class relations (ablation) | Negative prototype generator · FSOR | macro-averaged F1-score | Helped |
| Use task-adaptive negative generation rather than threshold-based classification | FSOR methods · Mini-ImageNet and Tiered-ImageNet | macro-averaged F1-score | Helped |
Closest papers by meaning
- Glocal Energy-Based Learning for Few-Shot Open-Set Recognition (CVPR 2023 · similarity 0.89)
- Few-Shot Open-Set Recognition by Transformation Consistency (CVPR 2021 · similarity 0.87)
- Meta Evidential Transformer for Few-Shot Open-Set Recognition (ICML 2024 · similarity 0.85)
- All Beings Are Equal in Open Set Recognition (AAAI 2024 · similarity 0.85)
- Frequency Shuffling and Enhancement for Open Set Recognition (AAAI 2024 · similarity 0.84)
- Few-Shot Recognition via Stage-Wise Retrieval-Augmented Finetuning (CVPR 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.