缩放定律论文趋势 2023–2026

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。

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 50 篇，第 1 / 1 页

1. CVPR 2026A Unified Framework for Knowledge Transfer in Bidirectional Model Scaling
2. CVPR 2026AnyLift Scaling Motion Reconstruction from Internet Videos via 2D Diffusion
3. CVPR 2026FFP-300K Scaling First-Frame Propagation for Generalizable Video Editing
4. CVPR 2026Iterative Closed-Loop Motion Synthesis for Scaling the Capabilities of Humanoid
5. CVPR 2026LaSM Layer-wise Scaling Mechanism for Defending Pop-up Attack on GUI
6. CVPR 2026MatAnyone 2 Scaling Video Matting via a Learned Quality Evaluator
7. CVPR 2026PointWorld Scaling 3D World Models for In-The-Wild Robotic Manipulation
8. CVPR 2026Scaling Dense Event-Stream Pretraining from Visual Foundation Models
9. CVPR 2026Scaling Multi-Identity Consistency for Image Customization via Multi-to-Multi Matching Paradigm
10. CVPR 2026Scaling Spatial Intelligence with Multimodal Foundation Models
11. CVPR 2026Scaling View Synthesis Transformers
12. CVPR 2026Scaling-Aware Data Selection for End-to-End Autonomous Driving Systems
13. CVPR 2026VibeToken Scaling 1D Image Tokenizers and Autoregressive Models for Dynamic
14. CVPR 2026Vocabulary Scaling Law Tuning Open-vocabulary Predictors for Their Openness
15. CVPR 2026WebGym Scaling Training Environments for Long-Horizon Visual Web Agents with
16. CVPR 2025Accurate Scene Text Recognition with Efficient Model Scaling and Cloze Self-Distillation
17. CVPR 2025Cubify Anything: Scaling Indoor 3D Object Detection
18. CVPR 2025D^3: Scaling Up Deepfake Detection by Learning from Discrepancy
19. CVPR 2025FASTer: Focal token Acquiring-and-Scaling Transformer for Long-term 3D Objection Detection
20. CVPR 2025Finsler Multi-Dimensional Scaling: Manifold Learning for Asymmetric Dimensionality Reduction and Embedding
21. CVPR 2025Flash3D: Super-scaling Point Transformers through Joint Hardware-Geometry Locality
22. CVPR 2025MegaSynth: Scaling Up 3D Scene Reconstruction with Synthesized Data
23. CVPR 2025Scaling Inference Time Compute for Diffusion Models
24. CVPR 2025Scaling Properties of Diffusion Models For Perceptual Tasks
25. CVPR 2025Scaling up Image Segmentation across Data and Tasks
26. CVPR 2025Scaling Vision Pre-Training to 4K Resolution
27. CVPR 2025ScaMo: Exploring the Scaling Law in Autoregressive Motion Generation Model
28. CVPR 2025SGSST: Scaling Gaussian Splatting Style Transfer
29. CVPR 2025Stretching Each Dollar: Diffusion Training from Scratch on a Micro-Budget
30. CVPR 2025T-CIL: Temperature Scaling using Adversarial Perturbation for Calibration in Class-Incremental Learning
31. CVPR 2025Towards Precise Scaling Laws for Video Diffusion Transformers
32. CVPR 2025UNIALIGN: Scaling Multimodal Alignment within One Unified Model
33. CVPR 2024Alpha Invariance: On Inverse Scaling Between Distance and Volume Density in Neural Radiance Fields
34. CVPR 2024An Empirical Study of Scaling Law for Scene Text Recognition
35. CVPR 2024DUDF: Differentiable Unsigned Distance Fields with Hyperbolic Scaling
36. CVPR 2024From Activation to Initialization: Scaling Insights for Optimizing Neural Fields
37. CVPR 2024InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks
38. CVPR 2024Scaling Laws for Data Filtering-- Data Curation cannot be Compute Agnostic
39. CVPR 2024Scaling Laws of Synthetic Images for Model Training ... for Now
40. CVPR 2024Scaling Up Dynamic Human-Scene Interaction Modeling
41. CVPR 2024Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild
42. CVPR 2024Scaling Up Video Summarization Pretraining with Large Language Models
43. CVPR 2023ConvNeXt V2: Co-Designing and Scaling ConvNets With Masked Autoencoders
44. CVPR 2023Galactic: Scaling End-to-End Reinforcement Learning for Rearrangement at 100k Steps-per-Second
45. CVPR 2023LargeKernel3D: Scaling Up Kernels in 3D Sparse CNNs
46. CVPR 2023On Data Scaling in Masked Image Modeling
47. CVPR 2023Reproducible Scaling Laws for Contrastive Language-Image Learning
48. CVPR 2023ResFormer: Scaling ViTs With Multi-Resolution Training
49. CVPR 2023Scaling Language-Image Pre-Training via Masking
50. CVPR 2023VideoMAE V2: Scaling Video Masked Autoencoders With Dual Masking

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。