Research map / Architectures and efficiency / Neural architecture search
One-shot and few-shot supernets
Trains supernets and ranks architectures with few-shot splitting and zero-cost proxies.
30 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2019–2025) · an approach in Neural architecture search, keyword cluster “search nas · bench · architecture search”
State History Benchmarks Tried Checklist Preprints Who Export
Where it stands
Few recent reports None of the tracked benchmarks has a report from 2025–26.
9 papers in 2023–24 and 6 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic held steady, from 12.0% to 12.0%.
No longer reported by papers here: ImageNet (Top-1 accuracy (%), last in 2023).
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
- 2019 · BayesNAS: A Bayesian Approach for Neural Architecture Search (ICML 2019 · 85 citations)
- 2020 · GreedyNAS: Towards Fast One-Shot NAS With Greedy Supernet (CVPR 2020 · 138 citations)
- 2021 · Zero-Cost Proxies for Lightweight NAS (ICLR 2021 · 65 citations)
- 2022 · NAS-Bench-Suite-Zero: Accelerating Research on Zero Cost Proxies (NeurIPS 2022 · 13 citations)
- 2023 · ZiCo: Zero-shot NAS via inverse Coefficient of Variation on Gradients (ICLR 2023 · 18 citations)
- 2024 · AZ-NAS: Assembling Zero-Cost Proxies for Network Architecture Search (CVPR 2024 · 27 citations)
- 2025 · ParZC: Parametric Zero-Cost Proxies for Efficient NAS (AAAI 2025 · 9 citations)
Earliest papers here
- BayesNAS: A Bayesian Approach for Neural Architecture Search (ICML 2019 · 85 citations)
- GreedyNAS: Towards Fast One-Shot NAS With Greedy Supernet (CVPR 2020 · 138 citations)
- NAS-Bench-1Shot1: Benchmarking and Dissecting One-shot Neural Architecture Search (ICLR 2020 · 77 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.
ImageNet · Top-1 accuracy (%) (higher is better)
5 papers, 2021–2023 · not reported since 2023
| Year | Best so far | Set by |
|---|---|---|
| 2021 | 80.5 | Few-Shot Neural Architecture Search (ICML 2021 · 9 citations) |
| 2023 | 82.6 | PreNAS: Preferred One-Shot Learning Towards Efficient Neural Architecture Search (ICML 2023 · 3 citations) |
What has been tried
28 papers here report 322 measured changes (a component added, removed or swapped, compared with a baseline): helped 70%, no clear effect 7%, hurt 20%, mixed 3%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| add trainability to the expressivity and progressivity proxies | CIFAR-10 · Kendall’s τ | 0.601 -> 0.679 | Helped | AZ-NAS: Assembling Zero-Cost Proxies for Network Architecture Search (CVPR 2024 · 27 citations) |
| add Mixer architecture to the predictor (vs MLP baseline) | NAS-Bench-201, 78 samples · KD (%) | 62.29 vs 54.64 (+7.65) | Helped | ParZC: Parametric Zero-Cost Proxies for Efficient NAS (AAAI 2025 · 9 citations) |
| Add CaLR without MS | CIFAR-10 (NAS-Bench-201) · Kendall's Tau | 0.751 -> 0.805 | Helped | Subnet-Aware Dynamic Supernet Training for Neural Architecture Search (CVPR 2025 · 4 citations) |
| Use sample-wise rather than standard activation patterns at input size 32×32 | NAS-Bench-201 CIFAR-10 · Spearman correlation | 0.34 -> 0.90 | Helped | SWAP-NAS: Sample-Wise Activation Patterns for Ultra-fast NAS (ICLR 2024 · 3 citations) |
| Add hierarchy-wise partition instead of edge-wise partition | CIFAR-10 · error rate | 2.86 -> 2.51 | Helped | HEP-NAS: Towards Efficient Few-shot Neural Architecture Search via Hierarchical Edge Partitioning (AAAI 2025 · 1 citation) |
| Remove the gradient mean from L-SWAG | NASBench-201 Cifar-10 · Spearman ρ | 0.75 -> 0.78 | Helped | L-SWAG: Layer-Sample Wise Activation with Gradients Information for Zero-Shot NAS on Vision Transformers (CVPR 2025 · 1 citation) |
| multiply V-PCA by parameter count to form W-PCA | FlexiBERT benchmark · Kendall τ | 0.466 -> 0.526 | Helped | W-PCA Based Gradient-Free Proxy for Efficient Search of Lightweight Language Models (ICLR 2025 · 0 citations) |
What reviewers will expect
Datasets and baselines named in the 6 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- CIFAR-10 · 4 of 6 papers
- CIFAR-100 · 4 of 6 papers
- NAS-Bench-201 · 4 of 6 papers
- ImageNet · 3 of 6 papers
- ImageNet16-120 · 3 of 6 papers
- NAS-Bench-101 · 2 of 6 papers
Baselines
- PC-DARTS · 2 of 6 papers
- Synflow · 2 of 6 papers
- Fisher · 2 of 6 papers
- NWOT · 2 of 6 papers
- ZiCo · 2 of 6 papers
- TF-TAS · 2 of 6 papers
- SPOS · 2 of 6 papers
- SWAP · 2 of 6 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 15 papers; from OpenAlex and the papers' affiliations).
- Hong Kong University of Science and Technology · 2 papers
- Kuaishou Technology · 2 papers
- Northeastern University · 2 papers
- Yonsei University · 2 papers
Use it
Download BibTeX of 25 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
- Efficient NAS search strategies (86 papers)
- Bi-level differentiable NAS (84 papers)
Elsewhere in the topic
- Latency-aware architecture search (91 papers)
- AutoML pipeline search (57 papers)
- Differentiable architecture search (DARTS) (32 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).