Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Meta-learning analysis and benchmarks
Analyzes meta-learned representations and builds benchmarks and Bayesian or convex-optimization learners.
50 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2018–2026) · an approach in Zero-/few-shot and meta-learning, keyword cluster “maml · meta algorithms · meta meta”
State History Benchmarks Tried Checklist Preprints Who Export
Where it stands
Quiet None of the tracked benchmarks has a report from 2025–26.
7 papers in 2023–24 and 4 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 3.4% to 2.3%.
No longer reported by papers here: miniImageNet 5-way 5-shot (accuracy, last in 2024); miniImageNet 5-way 1-shot (accuracy, last in 2024); tieredImageNet 5-way 5-shot (accuracy, last in 2024); CIFAR-FS 5-way 1-shot (accuracy (%), last in 2022); CIFAR-FS 5-way 5-shot (accuracy (%), last in 2022); Omniglot 20-way 1-shot (accuracy, last in 2022).
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2018 · Low-Shot Learning From Imaginary Data (CVPR 2018 · 772 citations)
- 2019 · Meta-Learning With Differentiable Convex Optimization (CVPR 2019 · 1,317 citations)
- 2020 · Few-shot Text Classification with Distributional Signatures (ICLR 2020 · 121 citations)
- 2021 · BOIL: Towards Representation Change for Few-shot Learning (ICLR 2021 · 60 citations)
- 2022 · Learning What Not To Segment: A New Perspective on Few-Shot Segmentation (CVPR 2022 · 330 citations)
- 2023 · Bi-Level Meta-Learning for Few-Shot Domain Generalization (CVPR 2023 · 36 citations)
- 2024 · Unleashing the Power of Meta-tuning for Few-shot Generalization Through Sparse Interpolated Experts (ICML 2024 · 0 citations)
- 2025 · Task-Specific Gradient Adaptation for Few-Shot One-Class Classification (CVPR 2025 · 0 citations)
- 2026 · Data-Centric Meta-Learning for Robust Few-Shot Generalization (CVPR 2026)
Earliest papers here
- Low-Shot Learning From Imaginary Data (CVPR 2018 · 772 citations)
- Meta-Learning With Differentiable Convex Optimization (CVPR 2019 · 1,317 citations)
- Meta-Transfer Learning for Few-Shot Learning (CVPR 2019 · 1,294 citations)
Benchmark records
The best number each year that beat every earlier one, among papers of this approach reporting their own method. Numbers are extracted by AI from each paper's text; check the paper before citing.
miniImageNet 5-way 5-shot · accuracy (higher is better)
10 papers, 2019–2024 · not reported since 2024
| Year | Best so far | Set by |
|---|---|---|
| 2019 | 80 | Meta-Learning With Differentiable Convex Optimization (CVPR 2019 · 1,317 citations) |
| 2022 | 86.12 | Improving Task-Specific Generalization in Few-Shot Learning via Adaptive Vicinal Risk Minimization (NeurIPS 2022 · 0 citations) |
miniImageNet 5-way 1-shot · accuracy (higher is better)
9 papers, 2019–2024 · not reported since 2024
| Year | Best so far | Set by |
|---|---|---|
| 2019 | 69.13 | LGM-Net: Learning to Generate Matching Networks for Few-Shot Learning (ICML 2019 · 50 citations) |
| 2022 | 74.63 | Improving Task-Specific Generalization in Few-Shot Learning via Adaptive Vicinal Risk Minimization (NeurIPS 2022 · 0 citations) |
tieredImageNet 5-way 5-shot · accuracy (higher is better)
5 papers, 2019–2024 · not reported since 2024
| Year | Best so far | Set by |
|---|---|---|
| 2019 | 81.75 | Meta-Learning With Differentiable Convex Optimization (CVPR 2019 · 1,317 citations) |
| 2022 | 86.06 | How to Train Your MAML to Excel in Few-Shot Classification (ICLR 2022 · 16 citations) |
CIFAR-FS 5-way 1-shot · accuracy (%) (higher is better)
4 papers, 2019–2022 · not reported since 2022
| Year | Best so far | Set by |
|---|---|---|
| 2019 | 72.8 | Meta-Learning With Differentiable Convex Optimization (CVPR 2019 · 1,317 citations) |
| 2022 | 82.02 | Improving Task-Specific Generalization in Few-Shot Learning via Adaptive Vicinal Risk Minimization (NeurIPS 2022 · 0 citations) |
CIFAR-FS 5-way 5-shot · accuracy (%) (higher is better)
4 papers, 2019–2022 · not reported since 2022
| Year | Best so far | Set by |
|---|---|---|
| 2019 | 85 | Meta-Learning With Differentiable Convex Optimization (CVPR 2019 · 1,317 citations) |
| 2022 | 88.25 | Improving Task-Specific Generalization in Few-Shot Learning via Adaptive Vicinal Risk Minimization (NeurIPS 2022 · 0 citations) |
Omniglot 20-way 1-shot · accuracy (higher is better)
4 papers, 2019–2022 · not reported since 2022
| Year | Best so far | Set by |
|---|---|---|
| 2019 | 97.2 | Adaptive Posterior Learning: few-shot learning with a surprise-based memory module (ICLR 2019 · 10 citations) |
What has been tried
45 papers here report 490 measured changes (a component added, removed or swapped, compared with a baseline): helped 68%, no clear effect 10%, hurt 17%, mixed 5%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| add adapter network A only | miniImageNet, Omniglot, and CIFAR-FS, 2-shot · accuracy | 69.1 -> 72.13; 96.6 -> 97.89; 70.0 -> 75.91 | Helped | Task-Specific Gradient Adaptation for Few-Shot One-Class Classification (CVPR 2025 · 0 citations) |
| add attentive neighbor aggregation | Nell-One 1-shot · MRR | 0.311 -> 0.322 | Helped | MoEMeta: Mixture-of-Experts Meta Learning for Few-Shot Relational Learning (NeurIPS 2025 · 0 citations) |
| Remove dense teachers | Meta-Dataset ID Avg · accuracy | 85.14 -> 85.07 | Hurt | Unleashing the Power of Meta-tuning for Few-shot Generalization Through Sparse Interpolated Experts (ICML 2024 · 0 citations) |
| Generate both KI and KD rather than only one of them | mini-ImageNet 5-way 5-shot · accuracy | 68.34±0.46% -> 72.02±0.44% | Helped | Enabling Few-Shot Learning with PID Control: A Layer Adaptive Optimizer (ICML 2024) |
What reviewers will expect
Datasets and baselines named in the 5 papers of this approach from 2024–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- Omniglot · 2 of 5 papers
- miniImageNet · 2 of 5 papers
- CIFAR-FS · 2 of 5 papers
Baselines
- MAML · 2 of 5 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 11 papers; from OpenAlex and the papers' affiliations).
- University of Chinese Academy of Sciences · 2 papers
- University of Edinburgh · 2 papers
- Microsoft · 2 papers
Use it
Download BibTeX of 26 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Meta-augmentation and meta-objectives (60 papers)
- Normalization and unsupervised meta-learning (39 papers)
Elsewhere in the topic
- Generative feature synthesis for zero-shot (117 papers)
- Zero-shot robustness and transfer (101 papers)
- Robust fine-tuning of zero-shot models (92 papers)
- Transductive and semantic-dictionary zero-shot (90 papers)
- Embedding adaptation for few-shot classification (71 papers)
- Fine-tuning strategies for few-shot (67 papers)
- Fine-grained and cross-domain few-shot baselines (59 papers)
- Alignment-based few-shot metric learning (38 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).