物理约束 ML论文趋势 2023–2026

主要结论

• 物理约束 ML论文从 2023 年的 50 篇增长到 2026 年的 199 篇，三年间扩大到约 4.0 倍。
• NeurIPS 是发物理约束 ML论文最多的会议，2023–2026 年共 167 篇，占四个会议总量的 35%。
• 交叉方向里涨得最快的是多模态大模型：同时属于两个主题的论文从 2023 年的 0 篇增至 2026 年的 11 篇。
• 在「气候与物理科学」的 5 个主题里，它的 2026 年论文数排第 1；同组增长最快的天气预报是 6.1 倍，物理约束 ML是 4.0 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 2026Derivative-Informed Training of Neural Operators On-the-Fly via Sketched Tangent Consistency
• ICML 2026Bias-Spectrum Neural Processes for Parametric PDEs: Architecture Priors Meet PDE Constraints
• ICLR 2026CONSIGN: Conformal Segmentation Informed by Spatial Groupings via Decomposition
• CVPR 2026Domain Sensitive Federated Learning with Fisher-Informed Pruning
• NeurIPS 2025Activation-Informed Merging of Large Language Models
• ICML 2025Geometry-Informed Neural Networks
• ICLR 2025Beyond the convexity assumption: Realistic tabular data generation under quantifier-free real linear constraints
• CVPR 2025A Physics-Informed Blur Learning Framework for Imaging Systems
• NeurIPS 2024ChaosBench: A Multi-Channel, Physics-Based Benchmark for Subseasonal-to-Seasonal Climate Prediction
• ICML 2024A Space Group Symmetry Informed Network for O(3) Equivariant Crystal Tensor Prediction
• ICLR 2024AirPhyNet: Harnessing Physics-Guided Neural Networks for Air Quality Prediction
• CVPR 2024A Physics-informed Low-rank Deep Neural Network for Blind and Universal Lens Aberration Correction
• NeurIPS 2023ClimSim: A large multi-scale dataset for hybrid physics-ML climate emulation
• ICML 2023Gradient Descent Finds the Global Optima of Two-Layer Physics-Informed Neural Networks
• ICLR 2023Bi-level Physics-Informed Neural Networks for PDE Constrained Optimization using Broyden's Hypergradients
• CVPR 20233D Human Pose Estimation via Intuitive Physics
• NeurIPS 2026Neural Modal Decomposition: Architectural Priors from Observables
• ICML 2026Fine-to-Coarse Fairness-Informed Multi-View Clustering
• ICLR 2026From Neural Networks to Logical Theories: The Correspondence between Fibring Modal Logics and Fibring Neural Networks
• CVPR 2026LaDy Lagrangian-Dynamic Informed Network for Skeleton-based Action Segmentation via Spatial-Temporal

全部物理约束 ML论文

共 481 篇，第 1 / 10 页

1. CVPR 2026A Geometric Algebra-Informed 3DGS Framework for Wireless Channel Prediction
2. CVPR 2026AeroAgent A Vision-Physics-Decision Framework for Aerodynamic Vehicle Design
3. CVPR 2026AnchorSplat Feed-Forward 3D Gaussian Splatting With 3D Geometric Priors
4. CVPR 2026AssemblyBench Physics-Aware Assembly of Complex Industrial Objects
5. CVPR 2026AviaSafe A Physics-Informed Data-Driven Model for Aviation Safety-Critical Cloud Forecasts
6. CVPR 2026Cell-Type Prototype-Informed Neural Network for Gene Expression Estimation from Pathology
7. CVPR 2026CICA Coupling Confidence-Aware Pretraining with Confidence-Informed Attention for Robust Multimodal
8. CVPR 2026Domain Sensitive Federated Learning with Fisher-Informed Pruning
9. CVPR 2026Dynamic Black-hole Emission Tomography with Physics-informed Neural Fields
10. CVPR 2026From 2D Alignment to 3D Plausibility Unifying Heterogeneous 2D Priors
11. CVPR 2026GaussFusion Improving 3D Reconstruction in the Wild with A Geometry-Informed
12. CVPR 2026GeoDexGrasp Geometry-aware Generation for Data-efficient and Physics-plausible Dexterous Grasping
13. CVPR 2026InterPhys Physics-aware Human Motion Synthesis in a Dynamic Scene
14. CVPR 2026LaDy Lagrangian-Dynamic Informed Network for Skeleton-based Action Segmentation via Spatial-Temporal
15. CVPR 2026Learning Surgical Robotic Manipulation with 3D Spatial Priors
16. CVPR 2026MaskAdapt Learning Flexible Motion Adaptation via Mask-Invariant Prior for Physics-Based
17. CVPR 2026MDS-VQA Model-Informed Data Selection for Video Quality Assessment
18. CVPR 2026MicroFM Physics-guided Flow Matching for Isotropic Microscopy Reconstruction
19. CVPR 2026PHASE-Net Physics-Grounded Harmonic Attention System for Efficient Remote Photoplethysmography Measurement
20. CVPR 2026PhyOceanCast Global Ocean Forecasting with Physics-Informed Diffusion
21. CVPR 2026Physics-Consistent Diffusion for Efficient Fluid Super-Resolution via Multiscale Residual Correction
22. CVPR 2026Physics-Guided Multistep Deformation Reversal for Ancient Bamboo Slip Restoration
23. CVPR 2026PhysSkin Real-Time and Generalizable Physics-Based Animation via Self-Supervised Neural Skinning
24. CVPR 2026PMRNet Physics-informed Multi-scale Refinement Network for Medical Image Segmentation
25. CVPR 2026Seeing through Light and Darkness Sensor-Physics Grounded Deblurring HDR NeRF
26. CVPR 2026Spatial-Spectral Residuals Informed Diffusion Neural Operator for Pan-sharpening
27. CVPR 2026TopoSlide Topologically-Informed Histopathology Whole Slide Image Representation Learning
28. CVPR 2026UniPixie Unified and Probabilistic 3D Physics Learning via Flow Matching
29. CVPR 2026VT-Intrinsic Physics-Based Decomposition of Reflectance and Shading using a Single
30. CVPR 2026WiseEdit Benchmarking Cognition- and Creativity-Informed Image Editing
31. ICLR 2026(U)NFV: Supervised and Unsupervised Neural Finite Volume Methods for Solving Hyperbolic PDEs
32. ICLR 2026**TandemFoilSet**: Datasets for Flow Field Prediction of Tandem-Airfoil Through the Reuse of Single Airfoils
33. ICLR 2026$\boldsymbol{\partial^\infty}$-Grid: A Neural Differential Equation Solver with Differentiable Feature Grids
34. ICLR 2026A Physics-Inspired Optimizer: Velocity Regularized Adam
35. ICLR 2026ASSESS: A Semantic and Structural Evaluation Framework for Statement Similarity
36. ICLR 2026ATEX-CF: Attack-Informed Counterfactual Explanations for Graph Neural Networks
37. ICLR 2026BioBO: Biology-informed Bayesian Optimization for Perturbation Design
38. ICLR 2026Bird's-eye-view Informed Reasoning Driver
39. ICLR 2026Boolean Satisfiability via Imitation Learning
40. ICLR 2026ComPhy: Composing Physical Models with end-to-end Alignment
41. ICLR 2026Compositional Generalization from Learned Skills via CoT Training: A Theoretical and Structural Analysis for Reasoning
42. ICLR 2026CONSIGN: Conformal Segmentation Informed by Spatial Groupings via Decomposition
43. ICLR 2026Converge Faster, Talk Less: Hessian-Informed Federated Zeroth-Order Optimization
44. ICLR 2026DeepPrim: a Physics-Driven 3D Short-term Weather Forecaster via Primitive Equation Learning
45. ICLR 2026DGNet: Learning Spatiotemporal PDEs with Discrete Green Networks
46. ICLR 2026DiffWind: Physics-Informed Differentiable Modeling of Wind-Driven Object Dynamics
47. ICLR 2026Distribution-informed Online Conformal Prediction
48. ICLR 2026Divide and Abstract: Autoformalization via Decomposition and Abstraction Learning
49. ICLR 2026DRIFT: Decompose, Retrieve, Illustrate, then Formalize Theorems
50. ICLR 2026Enhancing Stability of Physics-Informed Neural Network Training Through Saddle-Point Reformulation

常见问题

2023–2026 年四大会议一共有多少篇物理约束 ML论文？

共 481 篇，按年份分别是 2023 年 50 篇、2024 年 85 篇、2025 年 147 篇、2026 年 199 篇。

NeurIPS 2026 有多少篇物理约束 ML论文？

NeurIPS 2026 有 54 篇物理约束 ML论文，上一届是 65 篇。

ICLR 2026 有多少篇物理约束 ML论文？

ICLR 2026 有 64 篇物理约束 ML论文，上一届是 38 篇。

ICML 2026 有多少篇物理约束 ML论文？

ICML 2026 有 51 篇物理约束 ML论文，上一届是 25 篇。

CVPR 2026 有多少篇物理约束 ML论文？

CVPR 2026 有 30 篇物理约束 ML论文，上一届是 19 篇。

哪个会议的物理约束 ML论文最多？

NeurIPS，2023–2026 年共 167 篇。