Research map / Label-efficient and robust learning / Recognition and long-tail
Large-Scale Long-Tailed Recognition in an Open World
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 memory features and a concept selector to the direct feature (ablation) | OLTR · ImageNet-LT open-set, many-, medium-, and few-shot classes | top-1 accuracy | Helped |
| Add reachability-based confidence calibration (ablation) | OLTR · ImageNet-LT open-set few-shot classes | top-1 accuracy | Helped |
| Add modulated attention (ablation) | OLTR · ImageNet-LT open-set many-shot and few-shot classes | top-1 accuracy | Helped |
| Apply attention to self-attention maps instead of directly to feature maps (ablation) | OLTR attention module · ImageNet-LT open-set | top-1 accuracy | Helped |
| Add dynamic meta-embedding and modulated attention | Plain Model on MegaFace · MegaFace one-shot identities | identification rate: 84.60 -> 87.60 | Helped |
| Add dynamic meta-embedding and modulated attention | Plain Model on MegaFace · MegaFace zero-shot identities | identification rate: 77.72 -> 79.50 | Helped |
| Use OLTR rather than the plain model | ImageNet-LT many-shot classes · ImageNet-LT open-set | top-1 accuracy: 40.1 -> 41.9 | Helped |
| Use OLTR rather than the plain model | Places-LT medium-shot classes · Places-LT open-set | top-1 accuracy: 22.4 -> 36.8 | Helped |
Closest papers by meaning
- Bag of Tricks for Long-Tailed Visual Recognition with Deep Convolutional Neural Networks (AAAI 2021 · similarity 0.90)
- Global and Local Mixture Consistency Cumulative Learning for Long-Tailed Visual Recognitions (CVPR 2023 · similarity 0.90)
- Robust Visual Recognition with Class-Imbalanced Open-World Noisy Data (AAAI 2024 · similarity 0.89)
- SimLTD: Simple Supervised and Semi-Supervised Long-Tailed Object Detection (CVPR 2025 · similarity 0.89)
- Feature Fusion from Head to Tail for Long-Tailed Visual Recognition (AAAI 2024 · similarity 0.89)
- DiffuLT: Diffusion for Long-tail Recognition Without External Knowledge (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.