缩放定律论文趋势 2023–2026
数据/模型缩放
统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 605 篇被归入该主题的论文。
主要结论
- 缩放定律论文从 2023 年的 60 篇增长到 2026 年的 248 篇，三年间扩大到约 4.1 倍。
- NeurIPS 是发缩放定律论文最多的会议，2023–2026 年共 221 篇，占四个会议总量的 37%。
- 交叉方向里涨得最快的是时间序列：同时属于两个主题的论文从 2023 年的 1 篇增至 2026 年的 24 篇。
- 在「优化与缩放」的 5 个主题里，它的 2026 年论文数排第 5；同组增长最快的低秩适配是 3.0 倍，缩放定律是 4.1 倍。
各会议论文数
按年份和会议统计
| 年份 | NeurIPS | ICLR | ICML | CVPR | 合计 |
|---|---|---|---|---|---|
| 2023 | 21 | 14 | 17 | 8 | 60 |
| 2024 | 38 | 15 | 33 | 10 | 96 |
| 2025 | 80 | 48 | 56 | 17 | 201 |
| 2026 | 82 | 73 | 78 | 15 | 248 |
代表论文
按年份和会议均匀挑选，方便快速看到每一届的样貌
- NeurIPS 2026A Unified Merge Calculus for Learning-Rate Scaling on Neural Computation Graphs
- ICML 2026A Capacity-Based Rationale for Multi-Head Attention
- ICLR 2026$\mathbf{Li_2}$: A Framework on Dynamics of Feature Emergence and Delayed Generalization
- CVPR 2026A Unified Framework for Knowledge Transfer in Bidirectional Model Scaling
- NeurIPS 2025Audits Under Resource, Data, and Access Constraints: Scaling Laws For Less Discriminatory Alternatives
- ICML 2025A Simple Model of Inference Scaling Laws
- ICLR 2025(Mis)Fitting Scaling Laws: A Survey of Scaling Law Fitting Techniques in Deep Learning
- CVPR 2025D^3: Scaling Up Deepfake Detection by Learning from Discrepancy
- NeurIPS 2024An exactly solvable model for emergence and scaling laws in the multitask sparse parity problem
- ICML 2024A Dynamical Model of Neural Scaling Laws
- ICLR 2024Principled Architecture-aware Scaling of Hyperparameters
- CVPR 2024From Activation to Initialization: Scaling Insights for Optimizing Neural Fields
- NeurIPS 2023Abide by the law and follow the flow: conservation laws for gradient flows
- ICML 2023Dissecting the Effects of SGD Noise in Distinct Regimes of Deep Learning
- ICLR 2023Distributionally Robust Post-hoc Classifiers under Prior Shifts
- CVPR 2023LargeKernel3D: Scaling Up Kernels in 3D Sparse CNNs
- NeurIPS 2026ADIS-Law: Unified Scaling Laws for Annealing-Phase Domain Injection in Large Language Models
- ICML 2026Balancing Learning Rates Across Layers: Exact Two-Step Dynamics and Optimal Scaling in Linear Neural Networks
- ICLR 2026A universal compression theory: Lottery ticket hypothesis and superpolynomial scaling laws
- CVPR 2026MatAnyone 2 Scaling Video Matting via a Learned Quality Evaluator
全部缩放定律论文
共 605 篇，第 12 / 13 页
- CVPR 2023ResFormer: Scaling ViTs With Multi-Resolution Training
- CVPR 2023Scaling Language-Image Pre-Training via Masking
- CVPR 2023VideoMAE V2: Scaling Video Masked Autoencoders With Dual Masking
- ICLR 2023Block and Subword-Scaling Floating-Point (BSFP) : An Efficient Non-Uniform Quantization For Low Precision Inference
- ICLR 2023Broken Neural Scaling Laws
- ICLR 2023Distributionally Robust Post-hoc Classifiers under Prior Shifts
- ICLR 2023How Much Data Are Augmentations Worth? An Investigation into Scaling Laws, Invariance, and Implicit Regularization
- ICLR 2023Massively Scaling Heteroscedastic Classifiers
- ICLR 2023More ConvNets in the 2020s: Scaling up Kernels Beyond 51x51 using Sparsity
- ICLR 2023Proto-Value Networks: Scaling Representation Learning with Auxiliary Tasks
- ICLR 2023Reparameterization through Spatial Gradient Scaling
- ICLR 2023Scaling Forward Gradient With Local Losses
- ICLR 2023Scaling Laws For Deep Learning Based Image Reconstruction
- ICLR 2023Scaling up and Stabilizing Differentiable Planning with Implicit Differentiation
- ICLR 2023Scaling Up Probabilistic Circuits by Latent Variable Distillation
- ICLR 2023The Onset of Variance-Limited Behavior for Networks in the Lazy and Rich Regimes
- ICLR 2023Unsupervised Manifold Alignment with Joint Multidimensional Scaling
- ICML 2023Beyond the Universal Law of Robustness: Sharper Laws for Random Features and Neural Tangent Kernels
- ICML 2023Data Efficient Neural Scaling Law via Model Reusing
- ICML 2023Dissecting the Effects of SGD Noise in Distinct Regimes of Deep Learning
- ICML 2023Hidden Symmetries of ReLU Networks
- ICML 2023Learning Physical Models that Can Respect Conservation Laws
- ICML 2023Long Horizon Temperature Scaling
- ICML 2023On the Convergence Rate of Gaussianization with Random Rotations
- ICML 2023Online Platt Scaling with Calibeating
- ICML 2023Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling
- ICML 2023Scaling Laws for Generative Mixed-Modal Language Models
- ICML 2023Scaling Laws for Multilingual Neural Machine Translation
- ICML 2023Scaling Laws for Reward Model Overoptimization
- ICML 2023Scaling of Class-wise Training Losses for Post-hoc Calibration
- ICML 2023Scaling Spherical CNNs
- ICML 2023Scaling Up Dataset Distillation to ImageNet-1K with Constant Memory
- ICML 2023TAN Without a Burn: Scaling Laws of DP-SGD
- ICML 2023The case for 4-bit precision: k-bit Inference Scaling Laws
- NeurIPS 2023A Fast and Accurate Estimator for Large Scale Linear Model via Data Averaging
- NeurIPS 2023Abide by the law and follow the flow: conservation laws for gradient flows
- NeurIPS 2023DynaDojo: An Extensible Benchmarking Platform for Scalable Dynamical System Identification
- NeurIPS 2023Energy-Efficient Scheduling with Predictions
- NeurIPS 2023Focused Transformer: Contrastive Training for Context Scaling
- NeurIPS 2023Getting ViT in Shape: Scaling Laws for Compute-Optimal Model Design
- NeurIPS 2023Hierarchical clustering with dot products recovers hidden tree structure
- NeurIPS 2023Improved Bayes Risk Can Yield Reduced Social Welfare Under Competition
- NeurIPS 2023Neural Data Transformer 2: Multi-context Pretraining for Neural Spiking Activity
- NeurIPS 2023Performance Scaling via Optimal Transport: Enabling Data Selection from Partially Revealed Sources
- NeurIPS 2023R-divergence for Estimating Model-oriented Distribution Discrepancy
- NeurIPS 2023SAMRS: Scaling-up Remote Sensing Segmentation Dataset with Segment Anything Model
- NeurIPS 2023ScaleLong: Towards More Stable Training of Diffusion Model via Scaling Network Long Skip Connection
- NeurIPS 2023Scaling Data-Constrained Language Models
- NeurIPS 2023Scaling laws for language encoding models in fMRI
- NeurIPS 2023Scaling MLPs: A Tale of Inductive Bias
在 Claude、Cursor 里直接问这个趋势
接上 Paperlist MCP 后，写综述、挑方向或准备投稿时，不用离开对话框就能拿到这页的数字：每年各会议的论文数、增长倍数、涨得最快的子方向和代表论文，AI 会按你的问题整理。
可以直接复制这句话问 AI
用 paperlist 查缩放定律在 NeurIPS 2023–2026 的论文趋势，哪一年涨得最快？
AI 会调用的工具：get_topic_stats {"topic": "scaling-laws"}
常见问题
- 2023–2026 年四大会议一共有多少篇缩放定律论文？
- 共 605 篇，按年份分别是 2023 年 60 篇、2024 年 96 篇、2025 年 201 篇、2026 年 248 篇。
- NeurIPS 2026 有多少篇缩放定律论文？
- NeurIPS 2026 有 82 篇缩放定律论文，上一届是 80 篇。
- ICLR 2026 有多少篇缩放定律论文？
- ICLR 2026 有 73 篇缩放定律论文，上一届是 48 篇。
- ICML 2026 有多少篇缩放定律论文？
- ICML 2026 有 78 篇缩放定律论文，上一届是 56 篇。
- CVPR 2026 有多少篇缩放定律论文？
- CVPR 2026 有 15 篇缩放定律论文，上一届是 17 篇。
- 哪个会议的缩放定律论文最多？
- NeurIPS，2023–2026 年共 221 篇。