Research map / Architectures and efficiency
Neural architecture search: research map
548 papers on Neural architecture search in Architectures and efficiency, from ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR (2016–2026), grouped into 2 clusters and 6 approaches. The busiest year so far is 2021.
What it is
Neural architecture search (NAS) automates the design of network architectures by searching over candidates, often under a latency or size budget. Weight-sharing supernets and differentiable search (DARTS) made the search affordable.
Terms you will see on this page:
- NAS
- automatically searching for a good network architecture.
- Supernet
- one large network whose sub-networks share weights, so candidates need not be trained separately.
- DARTS
- differentiable architecture search, which relaxes the discrete choice so it can use gradients.
Start reading with the most cited papers here:
- Learning Transferable Architectures for Scalable Image Recognition (CVPR 2018 · 6,603 citations)
- Neural Architecture Search with Reinforcement Learning (ICLR 2017 · 4,370 citations)
Within Architectures and efficiency, its share shrank from 3.8% in 2023–24 to 1.4% in 2025–26 (75 → 50 papers at ICML, NeurIPS, ICLR, CVPR and AAAI, the venues with data for all four years).
Explore Neural architecture search in the interactive map
Working on something in this topic? Describe your idea in scime atlas to see which approach it falls under, the closest papers by meaning and how crowded the spot has become.
What changed recently
Approaches whose share of the topic moved most between 2023–24 and 2025–26 (ICML, NeurIPS, ICLR, CVPR and AAAI):
- Efficient NAS search strategies rising: 12% → 16% of the topic (9 → 8 papers)
- Latency-aware architecture search falling: 13% → 10% of the topic (10 → 5 papers)
Papers from 2025–2026 that are already cited most:
- ParZC: Parametric Zero-Cost Proxies for Efficient NAS (AAAI 2025 · 9 citations)
- Efficient Few-Shot Neural Architecture Search by Counting the Number of Nonlinear Functions (AAAI 2025 · 3 citations)
- Training-free Neural Architecture Search through Variance of Knowledge of Deep Network Weights (CVPR 2025 · 2 citations)
Approaches and key papers
Efficient NAS strategies · 337 papers
Keywords: search nas, bench, architecture search
Approaches in this cluster:
- Efficient NAS search strategies (130 papers)
Improves architecture search with credit assignment, distillation, and generative modeling of architectures. - Bi-level differentiable NAS (153 papers)
Reformulates gradient-based architecture search optimization with sampling, mixed-level, and geometry-aware methods. - One-shot and few-shot supernets (54 papers)
Trains supernets and ranks architectures with few-shot splitting and zero-cost proxies.
Most cited and most cited since 2024:
- Understanding and Simplifying One-Shot Architecture Search (ICML 2018 · 528 citations)
- CARS: Continuous Evolution for Efficient Neural Architecture Search (CVPR 2020 · 279 citations)
- AZ-NAS: Assembling Zero-Cost Proxies for Network Architecture Search (CVPR 2024 · 27 citations)
- DC-NAS: Divide-and-Conquer Neural Architecture Search for Multi-Modal Classification (AAAI 2024 · 21 citations)
Latency-aware and differentiable search · 211 papers
Keywords: differentiable, image, latency
Approaches in this cluster:
- Latency-aware architecture search (112 papers)
Searches network width and structure under latency, memory, and robustness constraints. - AutoML pipeline search (66 papers)
Searches model pipelines and operations beyond standard networks, including multimodal fusion and learning-curve ranking. - Differentiable architecture search (DARTS) (33 papers)
Relaxes architecture choice to continuous weights, improving DARTS robustness, memory use, and analysis.
Most cited and most cited since 2024:
- Learning Transferable Architectures for Scalable Image Recognition (CVPR 2018 · 6,603 citations)
- Neural Architecture Search with Reinforcement Learning (ICLR 2017 · 4,370 citations)
- Training-Free Quantum Architecture Search (AAAI 2024 · 26 citations)
- SasWOT: Real-Time Semantic Segmentation Architecture Search WithOut Training (AAAI 2024 · 23 citations)
Related topics in Architectures and efficiency
- Pruning and efficient tuning (1,335)
- Attention and transformers (2,598)
- Quantization (785)
- RNNs and normalization (3,071)
- Spiking networks (333)
- Mixture of experts (318)
- CNN design (1,051)
